Liquid cooling leak detection system

Abstract

The present disclosure relates to liquid cooling leak detection systems. The systems and methods herein are for leak detection in a computing environment using a shaped leak sensor. The shaped leak sensor can include an insulating material, a plurality of conductive traces printed or applied on a first side, and an adhesive on a second side. The shaped leak sensor can be configured to be shaped to at least match a layout around a component in the computing environment. A detector can monitor input from the shaped leak sensor to determine one or more of different states associated with the shaped leak sensor.

Description

Technical Field

[0001] At least one embodiment relates to cooling in computer environments, such as data centers. Background Technology

[0002] Computer environments, such as data centers, may employ liquid cooling technology. Liquid cooling connects to the computing components of a computer module via cold plates. While liquid cooling enhances heat dissipation from computing components, thus improving performance, it carries the risk of leakage. Therefore, leak detection ropes can be used for leak detection; when leaking liquid comes into contact with the rope, it triggers an alarm. However, leak detection ropes may lack flexibility and cannot conform perfectly to surfaces where leaking liquid may accumulate or be easily accessible. For example, in areas of a computer environment prone to liquid cooling loop leaks (such as around critical processors and memory components), leak detection ropes or other such leak detection components may be difficult to deploy. Furthermore, the detection surface area of ​​leak detection ropes or other such leak detection components (including flat rectangular designs) may be limited, and their bending radius may be restricted. A limited bending radius means that such leak detection components may be unable to detect leaks at the corners of computer modules. Given all these drawbacks, leak detection ropes and other such leak detection components may lack cost-effectiveness. Other solid-state sensors also have limitations in detecting liquid cooling loop leaks. Attached Figure Description

[0003] Figure 1 This is an illustration of a data center employing a shaped leak sensor in at least one embodiment;

[0004] Figure 2A This is an illustration of a computer module aspect of a shaped leakage sensor in at least one embodiment;

[0005] Figure 2B This is an illustration of an exemplary computer module having a region for forming a leak sensor in at least one embodiment;

[0006] Figure 2C This is an illustration of interconnected molded leak sensors disposed within a computer module in at least one embodiment;

[0007] Figure 2D This is a diagram illustrating the layout of the shaped leak sensor in the computer module in at least one embodiment;

[0008] Figure 2E This is a diagram illustrating aspects of the manufacturing of a shaped leak sensor in a computer module in at least one embodiment;

[0009] Figure 3AThis is an illustration of the forming aspects of providing a shaped leak sensor in at least one embodiment;

[0010] Figure 3B It is a circuit diagram of a detector associated with one or more shaped leak sensors in at least one embodiment;

[0011] Figure 3C This is a diagram illustrating the state that can be obtained using a detector associated with one or more shaped leak sensors in at least one embodiment;

[0012] Figure 3D This is an illustration of the use of a shaped leak sensor in a computing or switch tray in at least one embodiment;

[0013] Figure 4 The illustration shows a rack aspect in a system employing a shaped leak sensor according to at least one embodiment;

[0014] Figure 5A The process flow diagram illustrates a system with a shaped leak sensor in at least one embodiment;

[0015] Figure 5B The illustration shows another process flow of a system having a shaped leak sensor in at least one embodiment;

[0016] Figure 6A An exemplary data center is illustrated, in which it can be used Figures 1 to 5B At least one embodiment of the above;

[0017] Figure 6B This is a schematic diagram illustrating a computing system that can serve as a data center or high-performance computing (HPC) cluster, in which... Figures 1 to 5B At least one embodiment of the above;

[0018] Figure 6C The illustration depicts an exemplary computing environment in which [the following can be used] Figures 1 to 5B At least one embodiment of the above; and

[0019] Figure 6D The illustration depicts a computer system according to at least one example, in which... Figures 1 to 5B At least one embodiment of the above. Detailed Implementation

[0020] Figure 1This illustration depicts a data center employing a shaped leak sensor with a flexible insulating material in at least one embodiment. To address the limitations of liquid cooling technology in data centers, this document provides a system and method for a shaped leak sensor that can be fitted to the geometry around components and features in a computer environment and can be communicatively coupled together to extend leak detection capabilities. As used herein, a shaped leak sensor may refer to an insulating material that includes a leak-sensing component and incorporates a shape or formability suitable for an application, such as the geometry around components and features in a computer environment. The shaped leak sensor may be made of an insulating material, such as a flexible printed circuit board (or flex-PCB) with a polyimide substrate, and has gold (or other conductive materials, such as one or more of gold, graphite, silver, or nickel) traces printed or coated on one side to perform leak detection. Furthermore, the flex-PCB may be made of polyamide, polyester, polyethylene naphthalate, or polytetrafluoroethylene as exemplary substrates. The flexibility of the sensor allows it to move up and down and traverse gaps or surface height variations that may exist around components and features in a computer environment.

[0021] An adhesive can be applied to the other side of the flex-PCB. The flex-PCB may also include foam material around its perimeter to support the collection of any leaked fluid in the computing environment. Therefore, the shaped leak sensor can be shaped by cutting (e.g., with scissors) or any other suitable method to achieve mutual alignment; and the individual flex-PCBs can be plugged in and out via pin-and-socket connections without affecting their performance. The shaped leak sensor described herein may include parallel traces to support shaping by cutting the flex-PCB, thereby forming the shaped leak sensor. Furthermore, the shaped leak sensor can be configured for various applications in a computing environment through calibration to suit each individual application, and can also distinguish between different states. For example, this allows for very precise detection of specific small amounts of liquid in a leak.

[0022] This state can include: the absence / damage of a shaped leak sensor daisy chain coupled to different channels of an analog-to-digital converter (ADC) for voltage monitoring in leak detection; a shorted shaped leak sensor caused by improper contact of a metal component (in one example, damage from a computer environment); and different voltage levels for different possible levels of leakage. The ADC can be part of a detector, such as a board management controller (BMC) or other application-specific integrated circuit (ASIC), which uses a reference value in conjunction with the input from the shaped leak sensor to determine the state associated with the sensor. Therefore, this ADC does not rely on a hardwired comparator but can perform related processing using the reference value in conjunction with the input within the ADC's software or firmware. Finally, the shaped leak sensor can support reference value generation through calibration and can achieve testing functionality by changing the resistance between traces using field-effect transistors (FETs).

[0023] The shaped leak sensor of this application addresses the shortcomings of leak detection using leak detection ropes or other similar leak detection features, while also overcoming the limitations of solid-state sensors. These solid-state sensors not only have limitations in state detection but are also difficult to place in leak-prone areas of a computer environment (such as around critical processors and memory components). The shaped leak sensor based on flex-PCB can be shaped, fitted to irregular surfaces (such as above heat pipes and cold plates), and attached to other hard-to-reach areas and their surroundings in a computer environment; furthermore, it can monitor leaks in liquid cooling systems as well as other different states.

[0024] In one example, a computer module (which could be a server tray, switch tray, or switch box) may integrate a custom-shaped flex-PCB sensor to form the shaped leak sensor. Such shaped leak sensors can be fitted over the upper surface of a cold plate, heat pipe, or tray, positioned below or near all O-ring seals, hoses, manifolds, pipe fittings, and fluid couplings, or around the rack perimeter, as close as possible to all potential leak points. All shaped leak sensors used may include interlocked gold-plated traces with a predetermined spacing. In one example, this predetermined spacing may be 0.3 mm. Furthermore, when a conductive liquid, which may be part of a liquid-cooling circuit, contacts the shaped leak sensor, the traces and liquid can form a resistance that can be detected by a detector, which may be part of the sensing circuitry. The shaped leak sensor may terminate with a 1.1 MΩ resistor for presence detection. Furthermore, in one example, multiple (e.g., two) 2.2 megohm (MΩ) resistors can be provided, each belonging to a different molding leak sensor and connected in parallel as part of the molding leak sensor termination structure. Therefore, when a change in the measured resistance value is detected or it does not match the calibrated reference value, it can be determined that one of the molding leak sensors has been disconnected or is missing, thus identifying a specific molding leak sensor that has been disconnected or is missing.

[0025] Figure 1 This is a block diagram of an exemplary data center 100 having undergone the improved cooling system described in at least one embodiment. In at least one embodiment, the data center 100 may be equipped with a shaped leak sensor. The data center 100 may be one or more server rooms 102 having racks 110 and auxiliary equipment for housing one or more servers on one or more server trays having circuit boards, which are collectively referred to herein as computer modules. The data center 100 may be supported by a cooling tower 104 located outside the data center 100. The cooling tower 104 dissipates heat within the data center 100 by acting on a main cooling loop 106. Furthermore, a cooling distribution unit (CDU) 112 may be used between the main cooling loop 106 and a secondary cooling loop 108 to enable the extraction of heat from the secondary cooling loop 108 to the main cooling loop 106. In one aspect, the secondary cooling loop 108 may extend into the server trays via various piping systems as needed.

[0026] The primary cooling circuit 106 and secondary cooling circuit 108 are illustrated as line drawings, but those skilled in the art will recognize that one or more piping system features may be used. In one example, flexible polyvinyl chloride (PVC) pipes and associated piping systems may be used to deliver the medium in each of the primary cooling circuit 106 and secondary cooling circuit 108. In at least one embodiment, one or more pumps may be used to maintain the pressure differential within the primary cooling circuit 106 and secondary cooling circuit 108 to allow movement of the medium (such as a primary or secondary medium, which may be a coolant or refrigerant) based on temperature sensors at various locations (including within a server room, within one or more racks 110, and / or within server chassis or server trays within racks 110). As used herein, at least the secondary cooling circuit 108 associated with the primary cooling circuit 106 may be configured to use a cold plate to cool computing features of a computer module, and all such computing features and secondary cooling circuits are equipped with shaped leak sensors, details of which will be described herein. Figures 2A to 6D Further details are provided in one or more of the accompanying figures.

[0027] In at least one embodiment, the inlet temperature of the secondary medium in the secondary cooling circuit 108 is above 0 degrees Celsius (°C) but below 40°C, and it can exit at a temperature of approximately 60°C. In one example, the primary medium in the primary cooling circuit 106 can be used to cool the secondary medium in the secondary cooling circuit 108. The primary and secondary media may be at least water and additives, such as ethylene glycol or propylene glycol. In operation, each of the primary cooling circuit 106 and the secondary cooling circuit 108 has its own medium. In one aspect, the medium in the secondary cooling circuit may be dedicated to the requirements of components in the server tray or rack 110.

[0028] The CDU 112 may be capable of independent or concurrent complex control of the primary and secondary media in the primary cooling loop 106 and the secondary cooling loop 108. For example, the CDU may be adapted to control the flow rate of the secondary media in the secondary cooling loop 108, such that the secondary media can be appropriately distributed to extract heat generated within the rack 110. Furthermore, compared to the primary cooling loop, the secondary cooling loop 108 is equipped with more flexible piping 114 to connect to each computer module and provide secondary media to the computing features therein. In this disclosure, the term computing features is used interchangeably to refer to heat-generating components that benefit from this data center cooling system.

[0029] Figure 1 The pipe 118 shown, which may form part of the secondary cooling loop 108, can be referred to as a server room manifold. Additionally, an additional pipe 116 extending from this pipe 118 can also form part of the secondary cooling loop 108, but can be referred to as a row manifold. Furthermore, Figure 1The illustrated conduit 114 may enter the rack as part of the secondary cooling loop 108, but may be referred to as a rack cooling manifold. Furthermore, the manifold 116 may extend to all racks along a row in the data center 100. The piping system of the secondary cooling loop 108 (including manifolds or conduits 118, 116, and 114) may be improved through at least one embodiment of this disclosure. An optional cooler 120 may be located in the main cooling loop within the data center 102 to assist cooling before the cooling tower. If an additional loop exists in the main cooling loop, it will be understood by those skilled in the art, upon reading this disclosure, that such additional loop is used to provide cooling to the outside of the racks and the outside of the secondary cooling loop, and in this disclosure may be considered as the same loop as the main cooling loop.

[0030] In at least one embodiment, during operation, heat generated within the server tray of rack 110 can be transferred from at least one cold plate to the medium flowing out of rack 110 via the flexible conduit of the column manifold 114 of the secondary cooling circuit 108. Accordingly, secondary medium (in the secondary cooling circuit 108) from CDU 112 for cooling rack 110 flows to rack 110. Secondary medium from CDU 112 is transferred from one side of the server room manifold with conduit 118 via column manifold 116 to one side of rack 110, and passes through one side of the server tray via the provided conduit 114. Used secondary medium (or outflowing secondary medium carrying heat from computing features) can flow out from the other side of the server tray (e.g., into the left side of the rack and out from the right side of the server tray after circulating through the server tray or through components on the server tray). Used secondary media exiting the server tray or rack 110 flows out from different sides of conduit 114 (such as the outflow side) and moves to a parallel but also outflow side of the manifold 116. The used secondary media can move from the manifold 116 in the parallel portion of the data center manifold 118 in the opposite direction to the incoming secondary media (which could also be newer secondary media) and flow to the CDU 112. Furthermore, the used secondary media can have an outlet temperature above 0°C, specifically in the range of 40-60°C.

[0031] In at least one embodiment, the used secondary medium can exchange its heat with the primary medium in the main cooling circuit 106 via CDU 112. The used secondary medium can be renewed (e.g., relatively cooled compared to the temperature of the used secondary coolant stage) and prepared to be circulated back to the computing feature or component via the secondary cooling circuit 108. Various flow and temperature control features in CDU 112 are capable of controlling the heat exchanged from the used secondary medium or the flow rate of the secondary medium into and out of CDU 112. CDU 112 is also capable of controlling the flow of the primary medium in the main cooling circuit 106.

[0032] Figure 2AThis is an illustration of a computer module aspect of a shaped leak sensor in at least one embodiment. Aspect 200 may include server-level features and may include a computer module 202 having at least one server manifold 204 to allow cooling medium for the secondary cooling circuit 108 to enter and exit from the rack 110. However, the server manifold 204 may include separate channels for the medium inlet and outlet of the secondary cooling circuit 108, which is illustrated within the computer module as extending from the rack to form secondary cooling circuits 214A, 214B.

[0033] The secondary medium may enter from the rack manifold via inlet pipe 206 and exit via outlet pipe 208. On the server side, the secondary medium may flow via inlet pipe 210, through one or more cold plates 210A, 210B, and reach manifold 204 via outlet pipe 212. This indicates at least one or more secondary cooling circuits 214A, 214B within the server tray or case 202. These multiple secondary cooling circuits 214A, 214B may be extensions of the secondary cooling circuit 108 that interfaces with the primary cooling circuit 106, as they supply the same or substantially the same secondary medium from the secondary cooling circuit 108 to the cold plates 210A-210D. In at least one embodiment, the cold plates 210A-210D are associated with at least one computing component or feature 220A-220D. Furthermore, although illustrated as different cold plates, the cold plates 210A-210D shown can be part of a large, single cold plate structure with integrated contact points specifically located above the computing features 220A-220D located below. The computing features 220A-220D may include a processor, memory, and switches or regulators. In one example, the processor may include a graphics processing unit (GPU), a central processing unit (CPU), a data processing unit (DPU), a quantum processing unit (QPU), multiple parallel processing units (PPUs), and an ASIC.

[0034] In at least one embodiment, although inlet line 210 and outlet line 212 are illustrated as having one inlet and one outlet, multiple intermediate lines may be present, such as flexible tubes associating the cold plate with the respective inlet line 210 and outlet line 212. In at least one embodiment, an intermediate line directly coupling the cold plate to manifold 204 provides an inlet and outlet for such a connection. In at least one embodiment, a media adapter is provided to achieve such coupling. In at least one embodiment, the media adapter is sized to fit the inlet and outlet interfaces in the cold plate and manifold 204.

[0035] Figure 2AThe illustration also shows that the computer module aspect 200 may include a circuit board 222 having interconnect features 224 on a first side (the top side as shown) and a second side (the bottom side, having features similar to or soldered to the top side as shown). The interconnect features 224 may couple one or more computing features 220A-220D together. The interconnect features 224 may include copper traces, plated and unplated vias, solder joints, transmission lines, and electrically insulating circuit board material on which such copper traces and solder joints may be disposed.

[0036] In at least one implementation, the secondary cooling loops 108, 214A, and 214B can be used to capture most of the heat generated within the system, while simultaneously targeting the computing features 220A-220D. For example, ambient heat other than the target computing features 220A-220D can be captured. Therefore, one or more secondary cooling loops 108, 214A, and 214B can capture approximately 80% to 90% of the heat generated by the computer module or rack. This is true even though the secondary cooling loops 108, 214A, and 214B can operate at temperatures above 0°C, and even though the secondary cooling loops 108, 214A, and 214B can operate using a water-based medium. Figure 2A The overall illustration shows a computer module 202 with secondary cooling circuits 214A and 214B, which may include O-ring seals, hoses, manifolds, and fluid couplings. All these aspects of the secondary cooling circuits 214A and 214B can benefit from the shaped leak sensors described herein, aiming to be as close as possible to all potential leak locations.

[0037] Figure 2B This is an illustration of an exemplary computer module 230 having a region for forming a leak sensor in at least one embodiment. For example, Figure 2B The diagram illustrates the relationship between... Figure 2D A single-tray configuration with two trays. Exemplary computer module 230 may be... Figure 2AOne or more computer modules 202. Computer module 230 may include a server tray, compute tray, or server chassis having multiple shaped leak sensors among shaped leak sensors positioned below different aspects of secondary cooling loops 214A, 214B. The shaped leak sensors may be associated with the circuitry of detectors to detect leaks and notify cluster management via a baseboard management controller (BMC) in the data center. In at least one example, the detector may be a voltage-based leak detector or a capacitance-based leak detector. High-risk leak points may exist within computer module 230, and although quality and manufacturing processes may include stress testing of the completed cooling loop assembly to verify construction standards, all types of fluid connections may still pose a risk of leakage. Computer module 230 may have 20 or more potential leak locations, some of which are located around critical processor and memory components and have a tightbend radii.

[0038] exist Figure 2B The diagram illustrates at least one secondary cooling circuit 214A or 214B, having an inlet side 232A and an outlet side, discharge side, or outflow side 232B. Each of the inlet side 232A and the discharge side 232B may be associated with at least one adapter 234A, 234B, which may be coupled to a pipe, tube, or line 236. The inlet side 232A introduces a relatively cold medium to cool one or more computational features located below one or more cold plates 210A-210D. The medium may be distributed from the inlet side 232A and may return via the discharge side 232B. Distribution and return may be provided via at least one manifold 204, which may be used for the distribution side and the return side respectively, as... Figure 2B As shown.

[0039] Figure 2BThe illustration specifically illustrates multiple locations that may exist, some of which are roughly labeled 238A-238E, and these locations can be identified as potential leak areas. For example, the coupling of pipe, tube, or line 236 to adapters 234A, 234B can be identified as potential leak areas 238A, 238D. Adapters 234A, 234B may be quick-disconnect (QD) type adapters. The area associated with at least one manifold 204 can also be identified as potential leak areas 238B, 238C. Furthermore, the area associated with computing features (such as GPU, CPU, DPU, QPU, or PPU) and having a cold plate 210A-210B located above it can also be identified as a potential leak area 238E. All such areas may require shaped leak sensors to reduce the risk of potential leaks and also reduce the duration of potential leaks. In one example, coupled areas 238A and 238B may experience mating / non-matting connection cycles during a maintenance event and may be susceptible to a potential leak that may occur at some point and continue to occur after the maintenance event. All such areas may have a small bend radius, and portions of them may not be in contact with the leak detection rope.

[0040] While a leak tray can be provided above the power rack at the bottom of the rack, and this tray can guide leaking liquid through a drain hose to prevent such liquid from entering the high-pressure parts of the rack, it is beneficial to identify the leak immediately upon occurrence, or to identify the condition associated with a specific area of ​​the liquid cooling system to determine an appropriate response. Liquid in contact with the power rack or other power infrastructure can pose a hazard to personnel. Therefore, shaped leak sensors can be used in conjunction with such collection points, such as a leak bucket under the rack that collects liquid from drain pipes and vertical manifolds in the rack. Although Figure 2B The illustrations may be for computer modules 202 and 230, but the application of the forming leak sensor described herein is applicable to racks, at least as per this article. Figure 4 As described above. In one example, the leak bucket may have a spot leak sensor that can be associated as part of a daisy chain with a shaped leak sensor of the computer module, which can be integrally coupled to a rack infrastructure manager (RIM) or a BMS.

[0041] Figure 2CThis is an illustration of interconnected molding leak sensors 250 in at least one embodiment, which may be associated with each other and disposed within a computer module. The interconnected molding leak sensors 250 may include two or more of the following: CPU / DPU / QPU / PPU type molding leak sensor 260, one or more GPU type molding leak sensors 252A, 252B, adapter type molding leak sensor 254, and manifold type molding leak sensor 256. The interconnected molding leak sensors 250 may have a passive molding leak sensor, which can be connected to… Figure 2B The exemplary computer module 230 is inserted together with it. Regarding this document at least... Figure 2D This type of implementation is described in detail. Figure 2C The diagram also illustrates that each of the forming leak sensors 252A-260 may include multiple conductive traces 266 to achieve cutting 270 as part of the forming function of the forming leak sensor 252A-260.

[0042] As shown in the figure, conductive trace 266 extends from one conductor on one side and another conductor on the other side in an alternating finger configuration. The gaps between the alternating fingers allow for short circuits between the conductors, altering the resistance or capacitance of the shaped leak sensor. Furthermore, when the first side of the ring-shaped leak sensors 252A, 252B is cut 270, the integrity of the shaped leak sensor is maintained because the second side of the ring structure remains intact 272, allowing leak detection to be performed through the second side instead of the first side. For example, if a droplet or other medium causes contact between the alternating fingers of the conductors of conductive trace 266. In one example, this allows each shaped leak sensor to be precisely positioned in a small radius area and other areas around the component without interfering with the component.

[0043] In at least one example, the interconnected shaped leak sensors 250 can be interconnected via wires or cables 258A, 258B, which can be straight or Y-type cables. Wires or cables 258A, 258B can be coupled to detector 262, such as a voltage-based (V) detector or a capacitance-based detector. In at least one embodiment, detector 262 can be able to support any suitable communication protocol for communicating with the ADC, including... Internet Protocol (IP) and Communication protocol. The ADC can then use... or It communicates with the BMC, and the BMC can communicate with the Cluster Management System (CMS) via Ethernet. As shown in the figure, one or more shaped leak sensors are shaped to fit their application, conform to the surface in their application, and are located within corners with small bending radii.

[0044] Furthermore, at least one adapter-type molded leak sensor 254 among the interconnected molded leak sensors 250 can be positioned below a small leak collection tray within the computer modules 202, 230, enabling fluid coupling. However, in one example, the adapter-type molded leak sensor 254 may not be directly attached to the detector 262 or the cold plate. Instead, a Y-cable 258A can be provided to allow the adapter-type molded leak sensor 254 to be coupled to a manifold-type molded leak sensor 256, which is then coupled to the detector 262. In at least one example, such indirect coupling between the molded leak sensor 256 and the detector 262 may be based in part on a daisy-chain link between one molded leak sensor and another.

[0045] Furthermore, each of the interconnected molded leak sensors 250 can be molded in part based on an adhesive layer provided on at least one side to support peel-and-stick application, and molded to fit and mount onto its respective structure or surface. Each molded leak sensor may also include foam material 268 around its periphery to support the collection (as indicated by the arrow, for the movement of the medium collected here) of any fluid leaking in the computing environment. The matching respective structures or surfaces may represent the layout within computer modules 202, 230. Therefore, leak detection systems in computing environments (such as...) Figure 2C The interconnected shaped leak sensors 250 may include at least one shaped leak sensor among the interconnected shaped leak sensors 250. The at least one shaped leak sensor has a flexible insulating material with a plurality of conductive traces printed or applied on a first side and an adhesive on a second side. The shaped leak sensor may be configured to be shaped to at least match the layout around components in the computing environment, such as computing features, adapters, manifolds, tubes, etc.

[0046] Detector 262, acting as a voltage-based or capacitance-based leakage detector, can monitor input from the forming leakage sensor to determine one or more distinct states associated with the forming leakage sensor, as described herein at least regarding Figure 3B , Figure 3C As described in further detail, the insulating material used for the molded leak sensor can be a flexible insulating material and may include one or more of polyimide, polyamide, polyester, polyethylene naphthalate, or polytetrafluoroethylene. The conductive traces may include one or more of gold, graphite, silver, or nickel as at least one component of such materials used therein.

[0047] In at least one embodiment, the forming leak sensors can be daisy-chained or linked together via one or more suitable connectors that are compatible with each other. In one example, the CPU-type forming leak sensor 260 and one or more GPU-type forming leak sensors 252A, 252B can use a lockable flip-top zero-mating-force (ZIF) connector or... Connectors 264 are associated together. Connectors 264 as used herein may refer to one or more pins or sockets to allow shaped leak sensors to be associated with each other via a plug-in arrangement and directly or indirectly with detector 262. As used herein, pins or sockets may be conductive elements for electrical or communication connections and / or communication conductive elements that, once connected to another pin or socket, enable the transfer of power or communication between pins or sockets.

[0048] CPU-type forming leak sensor 260 and one or more GPU-type forming leak sensors 252A, 252B can communicate with detector 262 using a 2-pin coupling with a removable and reversible cable. Adapter-type forming leak sensor 254 and manifold-type forming leak sensor 256 can share a Y-cable 258A connected to detector 262. Furthermore, at least connector 264 determines the alignment of ring-type forming leak sensors 252A, 252B. While cut 270 can be used to align the ring-type forming leak sensor with the left or right side of the component (as per [reference]). Figure 3A (As further described), but the ring-shaped leak sensors are also aligned so that the cable does not need to extend around or interfere with the component. This is achieved by ensuring that connector 264 is aligned with other connectors of other ring-shaped leak sensors or with other connectors of other shaped leak sensors that are typically located in the same area.

[0049] Furthermore, the adapter-type shaped leak sensor 254 and the manifold-type shaped leak sensor 256 can be jointly associated with the Y-cable 258A via a welded connection. This is likely due to the low profile of the available space and the fact that the tray's fluid hose is directly above the potential leak area to be addressed. In addition to the welded connection, the other end of the Y-cable 258A can use the same connector as the CPU-type shaped leak sensor 260 and one or more GPU-type shaped leak sensors 252A, 252B. Figure 2CThe illustration also shows that the CPU-type shaped leak sensor 260 has a unique shape design that extends through conduits, adapters, or manifolds associated with the CPU cold plate 210C, and further to the GPU cold plate 210D. In at least one embodiment, the CPU-type shaped leak sensor 260 may be E-shaped and may be configured to connect to a motherboard associated with the detector 262 using a 2-pin cable 258B and a Molex connector.

[0050] Furthermore, GPU-type molded leak sensors 252A and 252B can be molded into a ring shape and can be connected to a centrally located molded leak sensor, such as CPU-type molded leak sensor 260. CPU-type molded leak sensor 260 and GPU-type molded leak sensors 252A and 252B can be connected in electrical parallel. Therefore, the presence of a molded leak sensor can be determined by evaluating the resistance associated with the leak detection system, as described herein. Figure 3B or Figure 3C As detailed in one or more of the above. For example, one or more shaped leak sensors and their associated cables may be terminated with a 2.2MΩ present resistor, and when all sensors are properly connected, the final parallel resistance will be exposed to the motherboard.

[0051] Figure 2D This is an illustration of the layout aspect 280 of a shaped leak sensor in a computer module in at least one embodiment. As shown, the shaped leak sensor can be configured to be flexible and have a small corner radius around computing features, adapters, manifolds, etc. Therefore, shaped leak sensors 252A-260 can be used to address one or more potential leak areas 238A-238E. Furthermore, the potential leak areas 238A-238E can be determined based on the layout information of the provided circuit board 222 and secondary cooling circuits 214A, 214B. Therefore, the shaped leak sensors 252A-260 can be designed to be mounted during the construction of one or more cold plates 210A-210D. Although the cold plates 210A-210D are illustrated as separate units, a large, monolithic cold plate integrating smaller cold plates 210A-210D can also be provided to cover one or more computing features 220A-220D shown in the accompanying drawings.

[0052] In at least one embodiment, information about Figure 6AThe computer system for calculating the features can be used with one or more pieces of first information associated with the first layout of circuit board 222 or the second layout of secondary cooling circuits 214A, 214D to generate second information associated with the design and interconnection of the shaped leak sensors 252A-260. Therefore, the shaped leak sensors 252A-260 described herein can be shaped to optimize the surface area to be covered, which can be the surface area other than the specific potential leak areas 238A-238E described herein. Furthermore, the interconnection design of the shaped leak sensors 252A-260 not only facilitates the inspection of such shaped leak sensors 252A-260 for damage, but also provides optimal assurance for the proper functioning of the leak detection function described herein from initial assembly to full tray integration. In one example, the shaped leak sensor 252A-260 can be designed as a single continuous shape that can both surround and adapt to a copper structure and be mounted under metal fittings and pipes on a cold plate.

[0053] Figure 2E This is an illustration of manufacturing aspect 290 of forming the leak sensor in at least one embodiment. Specifically, Figure 2E The illustration shows the layout of the circuit board and its provision. In the first information 292, molded leak sensors or interconnected molded leak sensors (dashed lines) can be designed as part of the second information 294 from the design subsystem 294A. This design can be provided to the manufacturing subsystem 294B to achieve at least the following: Figure 2C and Figure 2D The forming of the molded leak sensor or interconnected molded leak sensors is shown. Design subsystem 294A and manufacturing subsystem 294A can be part of a system of one or more circuits, which may include at least a processor and a memory having instructions to be executed by the processor to perform design and manufacturing functions. For example, design subsystem 294A may receive first information 292, and manufacturing subsystem 294B may use second information 296 to fabricate the molded leak sensor or interconnected molded leak sensor for the layout of circuit board 222. Therefore, one or more aspects 290 of a system having design subsystem 266 and manufacturing subsystem 268 can use... Figure 6A The data center 600 uses some or all of its features to perform the operation.

[0054] although Figure 2EAs illustrated, the second information 296 may be presented in specification or code form for printing, machining, or stamping features of at least the formed leak sensors or interconnected formed leak sensors described herein. For example, this specification or code may be used with other such manufacturing features of a 3D printer, a computer numerical control (CNC) machine tool, or manufacturing subsystem 294B to prepare one or more formed leak sensors or interconnected formed leak sensors.

[0055] Figure 3A This is an illustration of the forming aspect 300 for providing a formed leak sensor in at least one embodiment. For example, after the initial design is completed, if it is necessary to replace any of the continuous structures of the formed leak sensors 252A-260 shown in the figure, at least one side of the annular formed leak sensors 252A and 252B can be cut 270 and 302 to open up a portion of the annular formed leak sensors 252A and 252B, thereby enabling them to be fitted around the cold plate 210D. Since the other side 272 of the annular formed leak sensors 252A and 252B remains intact, the integrity of the annular formed leak sensors 252A and 252B can still be maintained as designed, and after installation around the component, the function of detecting leaks or other conditions can be achieved without disturbing the component. Therefore, when replacing the ring-shaped leak sensors 252A and 252B corresponding to the left GPU 220C, it is necessary to cut 270 a specific side of the ring-shaped leak sensor to achieve a fitting installation around the left GPU 220C shown in the figure; while when replacing the ring-shaped leak sensor corresponding to the right GPU 220A, it is necessary to cut the opposite side of the ring-shaped leak sensor to achieve a fitting installation around the right GPU 220A. However, the ring-shaped leak sensors 252A and 252B can be set without any specific mounting side bias, because each ring-shaped leak sensor 252A-260 can include multiple conductive traces 266 on the entire ring of the ring-shaped leak sensors 252A and 252B. In addition, the conductive traces 266 in each sensor can be an array of gold-plated conductive traces.

[0056] Although the shaped leak sensor described herein can operate normally without cutting except for the initial molding performed to adapt to the secondary cooling circuit design, a single-sided cutting scheme for the ring-shaped leak sensor is feasible at least, thanks to the continuity maintained on the other side of the ring-shaped leak sensor, thus maintaining the integrity of the leak detection system. Therefore, a single ring-shaped leak sensor can be provided, adapted to the left or right side of a computer module. Furthermore, if the cut 302 interferes with the left or right sides of the same ring-shaped leak sensor, a missing or open-circuit alarm will be triggered. Since the system described herein can also detect different states associated with the shaped leak sensor, such problems can be addressed immediately. Therefore, the multiple conductive traces 266 may include at least one pair of parallel conductive traces, which can be configured to be shaped to provide a shaped leak sensor. In one example, when at least one of the at least one pair of parallel conductive traces is shaped due to the performed cut 302, the parallel conductive trace can be configured to provide redundancy when sensing a leak.

[0057] Figure 3B This is an illustration of a circuit aspect 350 of a voltage-based leak detector associated with one or more shaped leak sensors in at least one embodiment. Circuit aspect 350 illustrates that each of the one or more shaped leak sensors can be enabled to provide a sensor alarm 352 to a corresponding channel 358 of the ADC 362. For each shaped leak sensor, one or more bias resistors and power supplies may be present. The ADC 362 may also be provided with a power supply voltage Vcc 354. However, a reference value or threshold 376 (in...) may be present. Figure 3C (In the middle), which may lie between a wider range of reference values ​​374A-374N. Such reference values ​​or thresholds 374A-374B, 376 may be retained by the ADC 362 in software or firmware during calibration of one or more molded leak sensors. One or more of the reference values ​​or thresholds 374A-374B, 376 may be applied to the values ​​received from the sensor alarm 352 in each channel 358; and may be used to determine a specific state 378A-378E that may be associated with each of the one or more molded leak sensors that provided the sensor alarm 352. The ADC 362 may provide information to the BMC 360 regarding the state of one or more molded leak sensors. Therefore, Figure 3B The hardware in the circuitry of the 350 can use an ADC 362 or a BMC 260 to set alarm thresholds in one or more shaped leak sensors.

[0058] In one example, when the resistance of the formed leak sensor exceeds the allowable range of a reference value or one of the thresholds 374A-374B, 376, an interrupt signal can be set and provided to the detector 262 associated with the BMC 360. In one example, this detector may involve one or more of the ADC 362 and the BMC 360. The ADC 362 may be able to aggregate all received interrupts and notify the BMC 360. The BMC 360 can communicate via integrated circuits. or The protocol accesses the ADC 362. However, the BMC 360 may also directly receive input from each of the forming leak sensors in the forming leak sensor suite. The BMC 360 can also determine when a read fails, which may indicate that a forming leak sensor has malfunctioned.

[0059] Furthermore, as the resistance associated with the forming leak sensor increases, the associated digital value (or digital readout) may fall within the higher range of the reference value or thresholds 374A-374B. The ADC 362 or BMC 360 can determine that the associated forming leak sensor may be disconnected or in a missing state (378A). This situation may be reported as a fault, either internally within the BMC 360 or transmitted from the ADC 362 to the BMC 360. When the resistance associated with the forming leak sensor decreases, the associated digital value may fall within the middle or lower range of the reference value or thresholds 374A-374B. The ADC 362 or BMC 360 can determine that a leak has occurred in a certain area or on a certain feature associated with the forming leak sensor. This leak may be a small leak (378C) or a large leak (378D). When the resistance associated with the forming leak sensor is zero or close to zero, a digital value may be the minimum of the reference value or thresholds 374A-374B. The ADC 362 or BMC 360 can determine that the molded leak sensor is in a short-circuit state 378E. In one example, when a metal fragment or component of the computer module 202 may become detached, it may come into contact with the molded leak sensor, potentially causing the molded leak sensor to be in a short-circuit state 378E. The ADC 362 or BMC 360 can also determine that the associated molded leak sensor is in a faulty state. Otherwise, the ADC 362 or BMC 360 can determine a normal state 378B for the molded leak sensor, which provides a voltage associated with a digital value within the normal ADC threshold 376.

[0060] In at least one embodiment, visual indications (such as colors or flashes emitted via light-emitting diodes (LEDs)) may be provided to display on the computer module where a leak exists, but connectors may also be used to provide this information to the BMC 360 or CMS 356. This connector may be an RJ-45 (or Ethernet) connector between at least the BMC 360 and CMS 356. All connectors associated with the leak detection system may be dedicated to leak detection itself and may provide leak detector indications on a tray or at the front of the computer module 202 or to one or more remote devices. Thus, although the shaped leak sensor and its associated integrated circuit (IC) may be located on the motherboard or circuit board 222, the leak detector indication information can support the determination of a leak condition or the status of the shaped leak sensor without contact with the circuit board 222.

[0061] Furthermore, leak detection systems can be calibrated and tested in part based on the type of medium used in the secondary cooling loop. For example, in When used as a cooling medium in a secondary cooling loop and monitored for leak detection, the PG25 can be used to pre-calibrate and test the leak detection system. For example, it can be determined that a drop of PG25 can cause a 2% decrease in the reading within the ADC 362. This decrease in reading may be relative to a resistor that can be configured in a leak detection system with a specific shaped leak sensor. However, a 2% decrease may indicate component tolerances of the shaped leak sensor. Therefore, to make the leak detection system resistant to false triggering interference, the threshold used in the system to detect the decrease in reading can be increased from 2% to 5%. In at least one embodiment, the threshold indicating a change in state associated with the shaped leak sensor can be between 5% and 16%. Accordingly, for PG25, this threshold can indicate a range from small to large leaks. Similarly, when deionized water (DI water) is used as the medium, a range of 10% to 39% can be used to indicate different degrees of leakage, from small to large leaks. Furthermore, the liquid type of the shaped leak sensor can be programmed at installation or operation, wherein the BMC can store and implement references or thresholds associated with the programmed liquid type for each state of the shaped leak sensor.

[0062] In at least one embodiment, to compensate for component variability, upon initial power-on, the leak detection system described herein may default to a dry-state computing environment and perform actual dry-state measurements on one or more shaped leak sensors as part of system calibration. The dry-state computing environment, as used herein, refers to a condition in which no liquid is present in the computing environment, which may be provided as part of testing or calibration, or generated by a leak within the computing environment. When the dry-state measurement results are determined to be within a reasonable range (e.g., an absolute range of 0.51–0.53), the system may construct a threshold or reference value, which is determined at least in part based on the aforementioned reasonable range.

[0063] Therefore, the shaped leak sensor of this paper can be configured for calibration or testing in one of the different applications in a computing environment. Furthermore, such calibration can be performed at least in a dry-state version of the computing environment to determine reference values ​​not only for the dry state. Calibration can be performed using one or more calibration modules 364. At least one aspect of one or more calibration modules 364 is to provide a dry state for calibration on the shaped leak sensor side. At least another aspect of one or more calibration modules 364 is to provide a reference or threshold for the ADC or BMC to process digital readings acquired in real time from the shaped leak sensor. For example, calibration can be provided for different states available in each shaped leak sensor by incorporating leaks, faults, and other different states. Detectors 262 (which may be one or more of ADC 362 or BMC 360) can be configured to distinguish different states 370, at least in part, based on different voltages (including voltage changes, such as voltage increases or decreases) from the circuit aspect 350, in conjunction with a calibrated reference or threshold in the leak detection system described herein.

[0064] Furthermore, the leak detection system described herein integrates a reference value or threshold reset function. This may be necessary in scenarios such as when replacing a molded leak sensor during maintenance or when the type of medium used changes. In at least one embodiment, the BMC 360 can be used to program and set a threshold or reference value for each molded leak sensor. Then, when the resistance value of a molded leak sensor exceeds the allowable range based on said threshold or reference value, the BMC 360 can be interrupted from performing any routine operations. The BMC 360 can be accessed via I... 2 The C interface queries the forming leak sensor that triggered the interrupt to receive further status information or other relevant information, and uses this information to determine the status associated with the forming leak sensor.

[0065] Figure 3CThis is an illustration of state 370 obtainable using a voltage-based leak detector associated with one or more shaped leak sensors in at least one embodiment. State 370 may be determined with reference to one or more reference values ​​or thresholds 376 associated with an ADC digital reading 372, which may be digital values ​​374A-374B generated by the ADC 362 in part based on voltage values ​​from the shaped leak sensors. The reference values ​​or thresholds 376 may reflect the upper and lower limits of the digital values ​​374A-374B shown for different states 378A-378E, thus it can be understood that the reference values ​​or thresholds 376 may correspond to sensor missing state 378A, small leak state 378C, large leak state 378D, and sensor short circuit state 378E.

[0066] For a dual-pallet configuration, multiple interconnected sets of forming leak sensors can exist. Therefore, each interconnected set of forming leak sensors can have or use its own ADC channel, and this arrangement requires at least two ADC channels. Furthermore, a dual-pallet configuration can result in four sensor groups. For example, a left manifold and a right manifold, as well as a left cold plate and a right cold plate, can be present. Figure 2B As shown, therefore, four different forming leak sensors can exist. Then, when a resistance change exceeding the nominal value occurs, the interconnected forming leak sensors can interrupt the BMC 360. The BMC 360 can read all available ADC values ​​(which can be four ADC values ​​from the four forming leak sensors). The BMC 360 can, in part, match the ADC digital readings 372 to a predefined or predetermined range based on calibration 364 for the four different forming leak sensors. As a result of the matching, the BMC 360 will be able to determine the state 370 of a specific forming leak sensor, including whether it is in a sensor-off state, normal state, small leak state, large leak state, or sensor-short-circuit state. The BMC 360 can trigger an event notification containing the fault condition corresponding to the associated state of a specific forming leak sensor in the interconnected group of forming leak sensors.

[0067] In at least one embodiment, the BMC 360 can be interrupted by either the ADC or any of the forming leak sensors. For example, to avoid polling the ADC from the BMC, a sensing integrated circuit (IC) can be used to generate an interrupt when a predefined or predetermined reference value or threshold 376 is exceeded. The ADC 362 can support a continuous monitoring mode that employs a window threshold that includes an upper and lower limit of the reference or threshold 376. For example, if one of the ADC digital readings 372 on one of the ADC channels 358 exceeds the window threshold range reflecting the ADC threshold 380, the ADC is used to assert an alarm.

[0068] The FPGA acts as the ADC 362 and receives sensor alarms 352 via a serial general-purpose input / output (SGPIO), routing the alarm to the BMC 360 to indicate that one or more ADC channels 358 have asserted an alarm. The BMC 360 can monitor the SGPIO individually. In one example, a leak detection alarm (LEAK DETECT ALERT) (low or high) can be asserted on the SGPIO. Upon assertion, the BMC 360 can query all ADC channels to determine which ADC channel has failed. The BMC 360 can then begin polling the ADCs to monitor whether a leak is developing and notify the event listener of any updates.

[0069] The leak detection system described herein is capable of performing tests involving leak self-testing. For example, leak detection and response can be tested via I / O expanders. A test module 366 can be used, which may include field-effect transistors (FETs) or other electrical components that can change the resistance, voltage, or capacitance in the leak detection system and can be removably coupled to one or more shaped leak sensors to simulate different states of the shaped leak sensors. For example, for testing purposes, the test module 366 can set the I / O output high and pull the sensor input low to simulate a leak scenario. In at least one embodiment, all service events can generate an event log or a locally or remotely recorded event log. At least a remote event log can be implemented via a Uniform Resource Identifier (URI) address for the remote event log. The event log may include a standard error message identifier (ID) and may include message parameters indicating one or more shaped leak sensors that can be associated with the event. In at least one embodiment, a backup shutdown timer configuration can be provided, allowing it to be configured with an expiration value. The BMC 360 can then shut down computer modules 202, 230 upon timer expiration and after a large leak is detected.

[0070] Figure 3DThis is an illustration of the use 380 of a shaped leak sensor in a compute or switch tray in at least one embodiment. Each compute or switch tray 384 may include a shaped leak sensor 382. The shaped leak sensor 382 may be adapted to other components associated with the compute or switch tray 384, including a circuit breaker 386A, a liquid valve 386B, a leak funnel 386, a drain hose 390E, and a power rack 388. The shaped leak sensor 382 may work in conjunction with other leak sensors, including a liquid supply leak rope sensor 386C, a ground leak rope sensor 390F, and a liquid collector and leak point sensor 390D. Use 380 may include a cluster shutdown protocol. For example, a rack shutdown may be performed once a leak is detected from any of the sensors 382, ​​386C, 390D, or 390F. When a small leak is detected, such as when detected by only a single compute or switch tray 384, the deployment of the cluster shutdown protocol may result in the next checkpoint being performed. When certain sensors detect a leak, such as the ground leak rope sensor 390F or the shaped leak sensor 382 of multiple computing or switch trays 384, a more substantial response may be performed than at the next checkpoint. A more substantial response may be required, at least in part, because of the potential for electric shock hazard in any aspect of the use of 380 as shown.

[0071] The system shown in 380 may require triggering an emergency shutdown, which is part of both the cluster shutdown protocol and a more substantial response measure. In one example, circuit breaker 386A can be used to disconnect rack-level power. One or more liquid valves 386B can be shut down. Cluster management software 390B can aggregate leakage signals from individual nodes representing compute or switch trays 384 in the rack. In one example, cluster management software 390B can provide one or more configuration options for aggregating leakage signals: send an alarm without performing any further action; send an alarm triggering software shutdown, followed by power isolation and liquid valve closure; and send a warning triggering immediate power isolation and liquid valve closure.

[0072] Configuration options can be specified for different severity categories. For example, a small leak from a single node might result in the execution of the next checkpoint in the configuration options. A large leak from a single node, or any type of leak from multiple nodes, might result in the execution of a more substantial response. A sensor failure might trigger a completely different response to address that sensor. The BMS 360 can be used to control circuit breakers 386A and liquid valves 386B. The cluster management module 390B and the BMS 360 can establish a communication path via the out-of-band management network 390A (OOB Comm.N / W) to execute one or more of the above aspects.

[0073] Figure 4 The illustration depicts a rack aspect of a system employing a shaped leak sensor according to at least one embodiment. Rack 402 has supports 404, 406 to suspend one or more cooling loop components within rack 402. In at least one embodiment, rack manifolds 412, 414 may be provided to guide media from the manifolds to a computer module 408 within rack 402. Rack manifold 412 allows media from a secondary cooling loop to flow out of the manifold, be conveyed via conduit 410 to a server tray or server chassis 408, flow out from outlet manifold 414, and return to the manifold via outlet conduit 412. The shaped leak sensor described herein can be applied to any of the illustrated server trays or server chassis constituting computer module 408; additional local distribution units can also provide gain if increased media flow pressure is required at any level of the rack. Therefore, although described herein with respect to a computer module, this leak detection system can still be applied to racks to detect leaks on mating surfaces within areas of small curvature radii of the rack. In such implementations, and in at least one example, leak detection can be monitored from within a rack-mounted computer module, which may have… Figure 3B The circuitry of the 350 includes the BMC and ADC sections.

[0074] Figure 5A The illustration depicts a process flow or method for a system having a shaped leak sensor in at least one embodiment. Method 500 may include: determining 502 a layout associated with a circuit board. This layout may be as targeted to… Figure 2E As described above. Method 500 may include: determining 504 the layout associated with the secondary cooling circuit. This may be the layout of the cold plate, adapter, piping, and manifold of the secondary cooling circuit. Method 500 may include: verifying or determining 506 that the layout of the shaped leak sensor has been determined at least based on the layout associated with the secondary cooling circuit. Method 500 may include: fabricating 508 a shaped leak sensor comprising an insulating material, a plurality of conductive traces printed or applied thereon on a first side, and an adhesive on a second side. Method 500 may include: making the shape in 510 of the shaped leak sensor at least match the layout around components in a computing environment. For example, the shape of a CPU-type shaped leak sensor may be shaped according to the layout of one or more of the cold plate, adapter, piping, and manifold of the secondary cooling circuit associated with the CPU. Method 500 may include: monitoring 512 the input from the shaped leak sensor using a detector to determine one or more distinct states associated with the shaped leak sensor. The detector may be a voltage-based leak detector.

[0075] Method 500 may include additional steps or sub-steps, wherein the insulating material is a flexible insulating material, including one or more of polyimide, polyamide, polyester, polyethylene naphthalate, or polytetrafluoroethylene. Furthermore, in method 500, the conductive traces may include one or more of gold, graphite, silver, or nickel. Method 500 may include additional steps or sub-steps, namely, providing foam or volumetric material around the formed leak sensor to support the collection of leaked fluid from the computing environment.

[0076] Method 500 may include additional steps or sub-steps enabling a shaped leak sensor to be associated with another shaped leak sensor via a plug-in arrangement using one or more pins or sockets. Method 500 may include further steps or sub-steps for establishing an association with a detector, which may be established directly or via another shaped leak sensor. Method 500 may include further steps or sub-steps for enabling at least one pair of parallel conductive traces among a plurality of conductive traces. The at least one pair of parallel conductive traces may be configured to be shaped to provide a shaped leak sensor; and redundancy is provided when a leak is sensed after at least one of the at least one pair of parallel conductive traces is shaped.

[0077] Figure 5B The illustration shows another process flow or method 550 for a system having a shaped leak sensor in at least one embodiment. Figure 5B Method 550 can be used alone, or as detailed. Figure 5A The further steps or sub-steps of method 500, with Figure 5A Method 500 is used in conjunction with the above. Method 550 may include: determining 552 the application of the shaped leak sensor. Method 550 may include: determining 554 the calibration or test to be performed on the shaped leak sensor. The determination of the calibration or test to be performed may be partly based on the application. Method 550 may include: preparing 556 the shaped leak sensor to support... Figure 5A Method 500 includes steps 508 and 510. Method 550 may include: verifying or determining that the 558 shaped leak sensor is ready for calibration or testing. Method 550 may include: calibrating or testing the 560 shaped leak sensor for the application, wherein calibration is performed at least in a dry-state version of the computing environment to determine reference values ​​for different states of the shaped leak sensor. Method 550 may include: configuring a 562 detector to distinguish different states associated with the leak detection sensor, and may be based in part on different voltages provided in the input relative to a reference value.

[0078] Figure 6A An exemplary data center 600 is illustrated, in which it can be used Figures 2A to 5BAt least one embodiment of the present invention. For example, an exemplary data center 600 may be used to support one or more preparation or activation steps for generating or providing a shaped leak sensor for at least one underlying component of the exemplary data center 600. However, in at least one embodiment, the data center 600 may also include a computer module equipped with a shaped leak sensor, as described herein. Figures 1 to 5B As stated above.

[0079] In at least one embodiment, the data center 600 includes a data center infrastructure layer 610, a framework layer 620, a software layer 630, and an application layer 640. In at least one embodiment, such as regarding... Figures 1 to 5B The features of the shaped leak sensor described herein may be implemented within or in collaboration with the exemplary data center 600. Furthermore, features for generating or providing the shaped leak sensor for at least one feature of a computer module or rack may be implemented within or in collaboration with the exemplary data center 600. In at least one embodiment, the infrastructure layer 610, frame layer 620, software layer 630, and application layer 640 may be provided, in part or in whole, via computing components on server trays located in rack 110 of data center 100. This enables the cooling system of this disclosure to efficiently and effectively direct cooling to certain computing features. Furthermore, various aspects of the data center (including the data center infrastructure layer 610, frame layer 620, software layer 630, and application layer 640) can be used to support the selection or design of the shaped leak sensor described herein, as referenced herein at least above. Figures 1 to 5B The subject of discussion. Therefore, regarding Figure 6A The discussion can be understood to apply, for example, to the hardware and software features required to enable or support the provision of shaped leak sensors.

[0080] In at least one embodiment, such as Figure 6A As shown, the data center infrastructure layer 610 may include a resource orchestrator 612, packet computing resources 614, and node computing resources (“nodes CR”) 616(1)-616(N), where “N” represents any positive integer. In at least one embodiment, nodes CR 616(1)-616(N) may include, but are not limited to, any number of central processing units (“CPUs”) or other processors (including accelerators, field-programmable gate arrays (FPGAs), graphics processing units, etc.), storage devices (such as dynamic read-only memory), storage devices (such as solid-state or disk drives), network input / output (“NWI / O”) devices, network switches, virtual machines (“VMs”), power modules, and cooling modules, etc. In at least one embodiment, one or more nodes CR 616(1)-616(N) may be servers having one or more of the aforementioned computing resources.

[0081] In at least one embodiment, packet computing resource 614 may include individual packets of node CRs housed within one or more racks (not shown), or multiple racks (also not shown) housed in data centers in different geographical locations. Individual packets of node CRs within packet computing resource 614 may include packet computing, networking, memory, or storage resources that can be configured or allocated to support one or more workloads. In at least one embodiment, several node computing resources, including CPUs or processors, may be combined within one or more racks to provide computing resources to support one or more workloads. In at least one embodiment, one or more racks may also include any number of power modules, cooling modules, and network switches, and these components may be arranged in any combination.

[0082] In at least one embodiment, resource orchestrator 612 may be configured or otherwise control one or more node computing resources 616(1)-616(N) and / or grouped computing resources 614. In at least one embodiment, resource orchestrator 612 may include a Software Design Infrastructure (“SDI”) management entity for data center 600. In at least one embodiment, resource orchestrator may include hardware, software, or some combination thereof.

[0083] In at least one embodiment, such as Figure 6A As shown, framework layer 620 includes a job scheduler 622, a configuration manager 624, a resource manager 626, and a distributed file system 628. In at least one embodiment, framework layer 620 may include a framework for supporting software 632 of software layer 630 and / or one or more applications 642 of application layer 640. In at least one embodiment, software 632 or application 642 may respectively include web-based service software or applications, such as those provided by Amazon Web Services, Google Cloud, and Microsoft Azure. In at least one embodiment, framework layer 620 may be, but is not limited to, a free and open-source software web application framework, such as Apache Spark. TM(Hereinafter referred to as "Spark"), which can utilize the distributed file system 628 for large-scale data processing (such as "big data"). In at least one embodiment, the job scheduler 622 may include a Spark driver to facilitate the scheduling of workloads supported by various layers of the data center 600. In at least one embodiment, the configuration manager 624 is capable of configuring different layers, such as the software layer 630 and the framework layer 620, including Spark and the distributed file system 628 for supporting large-scale data processing. In at least one embodiment, the resource manager 626 is capable of managing cluster or group computing resources mapped to or allocated to support the distributed file system 628 and the job scheduler 622. In at least one embodiment, the cluster or group computing resources may include group computing resources 614 at the data center infrastructure layer 610. In at least one embodiment, the resource manager 626 may coordinate with the resource orchestrator 612 to manage these mapped or allocated computing resources.

[0084] In at least one embodiment, the software 632 included in the software layer 630 may include software used in at least a portion of the node computing resources 616(1)-616(N), the group computing resources 614, and / or the distributed file system 628 of the framework layer 620. One or more types of software may include, but are not limited to, internet web search software, email virus scanning software, database software, and streaming video content software.

[0085] In at least one embodiment, the application 642 included in the application layer 640 may include one or more types of applications used for at least a portion of the node computing resources 616(1)-616(N), the group computing resources 614, and / or the distributed file system 628 of the framework layer 620. The one or more types of applications may include, but are not limited to, any number of genomics applications, cognitive computing, and machine learning applications, including training or inference software, machine learning framework software (such as PyTorch, TensorFlow, Caffe, etc.), or other machine learning applications used in combination with one or more embodiments.

[0086] In at least one embodiment, any of the configuration manager 624, resource manager 626, and resource orchestrator 612 can implement any number and type of self-modification actions based on any amount and type of data acquired in any technically feasible manner. In at least one embodiment, self-modification actions can alleviate the burden on the operator of data center 600 to make potentially poor configuration decisions and may prevent underutilized and / or poorly performing portions of the data center.

[0087] In at least one embodiment, data center 600 may include tools, services, software, or other resources to train one or more machine learning models or to predict or infer information using one or more machine learning models according to one or more embodiments described herein. In at least one embodiment, a machine learning model can be trained by computing weight parameters according to a neural network architecture using the software and computing resources described above regarding data center 600. In at least one embodiment, a trained machine learning model corresponding to one or more neural networks can be used to infer or predict information using the resources described above regarding data center 600 by using weight parameters computed through one or more training techniques described herein. Deep learning can be advanced using any suitable learning network and the computing power of data center 600. Therefore, hardware in the data center can be used to support deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs) simultaneously or in parallel. For example, once a network is trained and successfully evaluated to identify data within a subset or slice, the trained network can provide similar representative data for using the collected data.

[0088] In at least one embodiment, the data center 600 may use a CPU, application-specific integrated circuit (ASIC), GPU, DPU, QPU, PPU, FPGA, or other hardware to perform training and / or inference using the resources described above. A QPU is configured to perform one or more operations associated with a quantum algorithm. In some embodiments, each of the one or more QPUs may include multiple qubits, and the one or more QPUs may communicate with each other via quantum channels. In some embodiments, each of the multiple qubits may include local qubits, global qubits, and / or synchronization qubits. In some embodiments, the local qubits of each QPU may be configured to perform one or more operations associated with a quantum algorithm on the QPU to which it is associated. Furthermore, the one or more software and / or hardware resources described above may be configured as services to allow a user to train or perform information inference, such as pressure, flow rate, temperature, and location information, or other artificial intelligence services.

[0089] The inference and / or training logic 615 can be used to perform inference and / or training operations associated with one or more embodiments. In at least one embodiment, the inference and / or training logic 615 can be used to... Figure 6AIn the system, inference or prediction operations are performed based at least in part on weight parameters computed using the neural network training operations, neural network functions, and / or architectures or neural network use cases described herein. In at least one embodiment, the inference and / or training logic 615 may include, but is not limited to, hardware logic where computational resources are dedicated or otherwise specifically designed to combine weight values ​​or other information corresponding to one or more neuron layers within the neural network. In at least one embodiment, the inference and / or training logic 615 may be used in conjunction with an application-specific integrated circuit (ASIC), such as Google's... Processing unit, Graphcore TM Inference Processing Unit (IPU) or Intel Corporation's (Such as "Lake Crest") processors.

[0090] In at least one embodiment, the inference and / or training logic 615 may be used in conjunction with central processing unit (CPU) hardware, graphics processing unit (GPU) hardware, or other hardware such as a field-programmable gate array (FPGA). In at least one embodiment, the inference and / or training logic 615 includes, but is not limited to, code and / or data storage modules, which may be used to store code (such as graph code), weight values, and / or other information, including bias values, gradient information, momentum values, and / or other parameter or hyperparameter information. In at least one embodiment, each code and / or data storage module is associated with a dedicated computing resource. In at least one embodiment, the dedicated computing resource includes computing hardware that further includes one or more ALUs, which perform mathematical functions (such as linear algebra functions) only on the information stored in the code and / or data storage modules, and the results are stored in the active storage module of the inference and / or training logic 615.

[0091] As will be described below, numerous specific details will be set forth in order to provide a more complete understanding of at least one embodiment. However, it will be apparent to those skilled in the art that the inventive concept can be practiced without one or more of these specific details.

[0092] Figure 6B This is a schematic diagram illustrating a computing system 650 that can serve as a data center or high-performance computing (HPC) cluster, in which... Figures 1 to 5B At least one embodiment of the above. According to at least one embodiment, the computing system 650 may include multiple subsystems, such as multiple processing devices, multiple network devices, and multiple networks coupled to each other. The computing system 650 is designed with multiple integrated circuits (referred to as processing devices), wherein each integrated circuit may include one or more CPUs and GPUs, forming a powerful and flexible architecture.

[0093] Various processing devices are interconnected via NVLink high-speed interconnect technology or other high-speed interconnect methods to achieve high-speed communication between subsystems. Simultaneously, these processing devices are also connected via NICs or DPUs to ensure efficient data transmission within the computing system 650 and with one or more external networks 6530, 6536. In this example, system 650 includes a packet switch 6548 that connects NIC / DPU 6528 to network 6530, and a packet switch 6550 that connects NIC / DPU 6532 to network 6536.

[0094] Coupling the processing device via NVLink allows for seamless data exchange and parallel processing, thereby improving overall computing performance. The processing device connects to multiple networks via one or more network interface cards (NICs) or DPUs, enabling the system to handle complex multi-network tasks with high bandwidth and low latency. This configuration is ideal for demanding applications requiring significant processing power, such as artificial intelligence (AI), machine learning (ML), and data-intensive computing, while ensuring robust connectivity and scalability across a variety of network environments. The integrated circuits of the Computing System 650 may include one or more CPUs and one or more GPUs.

[0095] Figure 6B An exemplary architecture of a multi-GPU architecture is also illustrated. As shown in the figure, computing system 650 includes a processing device 6502 with a multi-GPU architecture. Specifically, processing device 6502 may be a system-on-a-chip and includes multiple subsystems such as CPU 6506, GPU 6508, and GPU 6510. CPU 6506 may be coupled to GPU 6508 via die-to-die (D2D) or chip-to-chip (C2C) interconnects 6512, such as ground reference signal interconnects (GRS interconnects). CPU 6506 may be coupled to GPU 6510 via D2D or C2C interconnects 6514. CPU 6506 may also be coupled to GPU 6508 and GPU 6510 via PCIe interconnects.

[0096] The CPU 6506 can be coupled to one or more NICs or DPUs, which in turn are coupled to one or more networks. For example, as Figure 6B As shown, CPU 6506 is coupled to a first NIC / DPU 6526, which is coupled to network 6530. CPU 6506 is also coupled to a second NIC / DPU 6528, which is coupled to network 6530 via switch 6548. For example, NIC / DPU 6526 and NIC / DPU 6528 can be coupled to network 6530 via Ethernet (ETH), NVLINK, or Infinite Bandwidth (IB) connections.

[0097] The computing system 650 also includes a processing device 6504 with a multi-GPU architecture. Specifically, the processing device 6504 includes multiple subsystems, including a CPU 6516, a GPU 6518, and a GPU 6520. The CPU 6516 can be coupled to the GPU 6518 via a D2D or C2C interconnect 6522. The CPU 6516 can be coupled to the GPU 6520 via a D2D or C2C interconnect 6524. The CPU 6516 can also be coupled to the GPU 6518 and GPU 6520 via a PCIe interconnect. The CPU 6516 can be coupled to one or more NICs or DPUs, which are coupled to one or more networks. For example, as... Figure 6B As shown, CPU 6516 is coupled to a first NIC / DPU 6532, which is coupled to network 6536. CPU 6516 is also coupled to a second NIC / DPU 6534, which is coupled to network 6536 via switch 6550. NIC / DPU 6532 and NIC / DPU 6534 can be coupled to network 6536 via Ethernet (ETH), NVLINK, or Infinite Bandwidth (IB) connections.

[0098] In at least one embodiment, processing device 6502 and processing device 6504 can communicate with each other via NIC / DPU 6538, such as via PCIe interconnect. Processing device 6502 and processing device 6504 can also communicate with each other via high-bandwidth communication interconnect 6540, such as NVLink interconnect or other high-speed interconnects. For example, Figure 6B The packet switches in the diagram can include Nvidia Quantum-2 switches. For example, the NIC / DPU in the diagram can include an Nvidia Bluefield DPU.

[0099] In various embodiments, any network device of computing system 650, such as any of NIC / DPU 6526, 6528, 6532, 6534 and 6538, and / or any of switches 6548 and 6550, may include a shaped leak sensor that can be matched to the geometry around components and features in computing system 650 and can be communicatively coupled together to extend leak detection capabilities.

[0100] Figure 6C An exemplary computing environment 670 is illustrated, in which the following can be used Figures 1 to 5BAt least one embodiment of the present disclosure is described. The exemplary computing environment 670 may include a shaped leak sensor that can be matched to the geometry around components and features in the exemplary computing environment 670 and can be communicatively coupled together to extend leak detection capabilities. It should be understood that embodiments of this disclosure can also be used with reference to alternative environments, and specific discussions of components are provided by way of non-limiting example and may include equivalents. Furthermore, various features have been removed for clarity and brevity. Additionally, the systems and methods can be used with a variety of different architectures. The exemplary computing environment 670 may include a server 672 that can be used to perform HPC workloads, such as AI training or machine learning model training. In one embodiment, server 672 may be an application instance or a compute node. Server 672 may include a CPU 674 associated with a switch 676, such as a Peripheral Component Interconnect Fast (PCIe) switch that controls at least some data transmissions through communication paths that interconnect various components. In one embodiment, CPU 674 may include a root complex processor.

[0101] PCIe switch 676 may also be associated with GPU 678 and DPU 680, and may transfer data between at least some of CPU 674, GPU 678, and DPU 680 and other components. In one embodiment, PCIe switch 676 may be associated with more than one GPU or more than one DPU. In another embodiment, PCIe switch 676 may be located within DPU 680. PCIe switch 676 may manage the transfer of at least some of the data between CPU 674, GPU 678, and DPU 680. In another embodiment, the number of GPUs associated with PCIe switch 676 may be equal to the number of DPUs associated with PCIe switch 676. In at least one embodiment, server 672 may include, but is not limited to, any combination of any number of CPUs 674, PCIe switch 676, GPUs 678, and / or DPUs 680. For example, in at least one embodiment, server 672 may include eight, sixteen, thirty-two, and / or more GPUs 678. In at least one embodiment, Figure 6C The communication paths for interconnecting various components (including but not limited to CPU 674, PCIe switch 676, GPU 678, and DPU 680) can be implemented using any suitable protocol, such as peripheral component interconnect (PCI) based protocols (e.g., PCIe), or other bus or point-to-point communication interfaces and / or protocols, such as NV-Link high-speed interconnect or interconnect protocols.

[0102] DPU 680 may include a network interface card (NIC) 682, DDR memory 680B, and a non-volatile memory fast (NVMe) device 680C. NIC 682 is capable of interfacing with network 684, which may also interfacing with additional NVMe devices available to DPU 680, such as via a fabric. In one embodiment, DPU 680 may not include NVMe device 680C. In another embodiment, NVMe device 680C may reside on server 672, rather than on DPU 680. In yet another embodiment, computing environment 670 may include more than one NVMe device 146, such as a first NVMe device in DPU 140 and a second first NVMe device on server 672 directly associated with PCIe switch 676. In one embodiment, DPU 680 may not include DDR memory 680B and may include compute storage service (CSS) in place of DDR memory 680B or include CSS in addition to DDR memory. For example, computing environment 670 may include DPU compute storage device (CS) memory 680D accessible to DPU 680 as part of the CS. Network 684 may be able to interface with DPU CS memory 680D via NIC 682 according to any suitable interface protocol, such as Ethernet Remote Direct Memory Access (RDMA), unlimited bandwidth, Fibre Channel, etc.

[0103] The total available memory for data storage in computing environment 670 can be expanded using DPU 680 on system nodes. DPU 680 can access memory pool 680A already available on server 672, such as Double Data Rate (DDR) memory, onboard NVMe devices, NVMe devices via the architecture, and CS. Memory pool 680A may include at least one of DDR memory 680B, NVMe devices 680C, and DPU CS memory 680D. DPU 680 may also be able to access available memory of other DPUs that are part of memory pool 680A, and other DPUs may be able to access available memory of DPU 680 (such as memory pool 680A). This available memory can be accessed and used for data storage without adding computing resources (such as compute nodes) that would be required by other solutions. The available memory pool 680A accessible to DPU 680 can be configured for server 672 to expand the total available memory for data storage, such as to reduce the data storage load on CPU 674 or GPU 678, thereby improving the utilization of its memory for processing. For example, during AI training, model states, residual states, activation functions, and checkpoints can be stored or unloaded onto a memory pool 680A accessible to the DPU 680.

[0104] Figure 6D The figure illustrates a computer system 690 according to at least one example, in which a computer system 690 can be used. Figures 1 to 5B At least one embodiment of the present disclosure. In at least one embodiment, the computer system 690 is configured to implement the various processes and methods described throughout the present disclosure.

[0105] In at least one embodiment, the computer system 690 includes (but is not limited to) at least one central processing unit (“CPU”) 6902 connected to a communication bus 6910 implemented using any suitable protocol, such as PCI (“Peripheral Component Interconnect”), Peripheral Component Interconnect Express (“PCI-Express”), AGP (“Accelerated Graphics Port”), HyperTransport, or any other bus or point-to-point communication protocol. In at least one embodiment, the computer system 690 includes (but is not limited to) main memory 6904, and control logic (e.g., implemented in hardware, software, or a combination thereof) and data are stored in the main memory 6904, which may take the form of random access memory (“RAM”). In at least one embodiment, a network interface subsystem (“network interface”) 6922 provides an interface to other computing devices and networks for receiving data from other systems and transmitting data to other systems from the computer system 690.

[0106] In at least one embodiment, the computer system 690 includes (but is not limited to) an input device 6908, a parallel processing system 6912, and a display device 6906, which may be implemented using conventional cathode ray tube (“CRT”), liquid crystal display (“LCD”), light-emitting diode (“LED”), plasma display, or other suitable display technologies. In at least one embodiment, user input is received from the input device 6908, such as a keyboard, mouse, touchpad, microphone, etc. In at least one embodiment, each of the foregoing modules may reside on a single semiconductor platform to form the processing system.

[0107] In at least one embodiment, a computer program in the form of machine-readable executable code or computer control logic algorithms is stored in main memory 6904 and / or secondary storage devices. If executed by one or more processors, the computer program enables system 690 to perform various functions according to at least one embodiment. Memory 6904, storage devices, and / or any other storage devices are possible examples of computer-readable media. In at least one embodiment, secondary storage devices may refer to any suitable storage device or system, such as hard disk drives and / or removable storage device drives, including floppy disk drives, magnetic tape drives, optical disk drives, digital universal optical disc (DVD) drives, recording devices, universal serial bus (USB) flash memory, etc. In at least one embodiment, the architecture and / or functionality of the foregoing figures can be implemented in the following scenarios: CPU 6902, parallel processing system 6912, integrated circuits having at least some of the functions of CPU 6902 and parallel processing system 6912, chipsets (e.g., a group of integrated circuits designed to work together and sold as a unit to perform related functions), and any suitable combination of integrated circuits.

[0108] In at least one embodiment, the architecture and / or functionality of the foregoing figures can be implemented in scenarios such as general-purpose computer systems, circuit board systems, game console systems designed for entertainment purposes, and dedicated systems. In at least one embodiment, the computer system 690 can take the form of a desktop computer, laptop computer, tablet computer, server, supercomputer, smartphone (e.g., wireless handheld device), personal digital assistant (PDA), digital camera, vehicle, head-mounted display, handheld electronic device, mobile phone device, television, workstation, game console, embedded system, and / or any other type of logic device.

[0109] In at least one embodiment, the parallel processing system 6912 includes, but is not limited to, multiple parallel processors (“PPUs”) 6914 and associated memory 6916. In at least one embodiment, the PPUs 6914 are connected to a host processor or other peripheral devices via interconnects 6918 and switches 6920 or multiplexers. In at least one embodiment, the parallel processing system 6912 distributes computational tasks among the parallelizable PPUs 6914—for example, as part of distributing computational tasks among multiple graphics processing units (“GPUs”) thread blocks. In at least one embodiment, although such shared memory may result in performance penalties compared to using local memory and registers residing in the PPUs 6914, the memory is shared and accessible (e.g., for read and / or write access) among some or all of the PPUs 6914. In at least one embodiment, the operation of the PPUs 6914 is synchronized using commands such as _syncthreads(), where all threads in a block (e.g., executing on multiple PPUs 6914) arrive at a point in code execution before continuing execution.

[0110] Other variations are within the spirit and scope of this disclosure. While various modifications and alternative constructions are possible with the disclosed technology, some exemplary embodiments have been shown in the accompanying drawings, and these embodiments have been described in detail above. However, it should be understood that this disclosure is not intended to be limited to the specific forms disclosed, but is intended to cover all modifications, alternative constructions, and equivalents falling within the spirit and scope of this disclosure as defined in the appended claims.

[0111] In describing the disclosed embodiments (especially in the context of the following claims), the terms “a,” “an,” and “the,” and similar designations, should be interpreted to cover both singular and plural forms, unless otherwise stated herein or explicitly contradicted by the context, and such terms are not limiting of the terminology. Unless otherwise stated, the terms “comprising,” “having,” “including,” and “containing” should be interpreted as open-ended terms (i.e., “including, but not limited to”). The term “connected,” when unmodified and referring to a physical connection, should be interpreted as partially or wholly contained in, attached to, or connected together, even in the presence of intervening elements. The descriptions of numerical ranges herein are intended only as a convenient way to refer to each individual value within the range, unless otherwise stated herein, and each individual value is incorporated into the specification as if it were individually enumerated herein. In at least one embodiment, unless otherwise stated or contradicted by the context, the terms “set” (e.g., “a group of items”) or “subset” should be interpreted as a non-empty set containing one or more members. Furthermore, unless otherwise stated or contradicted in the context, a “subset” of a corresponding set does not necessarily mean a proper subset of that corresponding set, and a subset can be equal to the corresponding set.

[0112] For connective expressions (e.g., phrases of the form "at least one A, B, and C" or "at least one A, B, and C"), unless explicitly stated otherwise or contradicted by the context, they should be understood in context as referring to any of A, B, and C, or any non-empty subset of the set consisting of A, B, and C. For example, in an exemplary example of a set containing three members, the connective phrases "at least one A, B, and C" and "at least one A, B, and C" refer to any of the following sets: {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C}. Therefore, such connective expressions are generally not intended to imply that some embodiments require at least one of each of A, B, and C. Furthermore, unless explicitly stated otherwise or contradicted by the context, the term "plurality" indicates a plural state (e.g., "plural items" refers to multiple items). In at least one embodiment, "plurality" refers to at least two items, but may be more than two if explicitly stated or implied by the context. Furthermore, unless otherwise stated or the context clearly indicates, the phrase “based on” means “at least partially based on”, not “based on only”.

[0113] The operations of the processes described herein may be performed in any suitable order unless otherwise stated herein or the context clearly contradicts it. In at least one embodiment, processes such as those described herein (or variations and / or combinations thereof) are executed under the control of one or more computer systems configured with executable instructions, and are executed jointly on one or more processors in the form of code (e.g., executable instructions, one or more computer programs, or one or more application programs), or implemented by hardware or a combination thereof. In at least one embodiment, the code is stored on a computer-readable storage medium, for example, in the form of a computer program containing a plurality of instructions executable by one or more processors.

[0114] In at least one embodiment, the computer-readable storage medium is a non-transient computer-readable storage medium that excludes transient signals (e.g., propagating transient electrical signals or electromagnetic transmission signals) but includes non-transient data storage circuitry (e.g., buffers, caches, and queues) within a transient signal transceiver. In at least one embodiment, code (e.g., executable code or source code) is stored on a set of one or more non-transient computer-readable storage media that store executable instructions (or other memory for storing executable instructions); when these executable instructions are executed by one or more processors of a computer system (i.e., the result of execution), the computer system will perform the operations described herein. In at least one embodiment, the set of non-transient computer-readable storage media includes multiple non-transient computer-readable storage media, and one or more individual non-transient storage media do not store all the code, but the multiple non-transient computer-readable storage media collectively store all the code. In at least one embodiment, the executable instructions are executed in such a manner that different instructions are executed by different processors. For example, instructions are stored on a non-transient computer-readable storage medium, a portion of which are executed by the main central processing unit (CPU), while the graphics processing unit (GPU) executes the remaining instructions. In at least one embodiment, different components of the computer system have separate processors, and different processors execute different subsets of instructions.

[0115] In at least one embodiment, the arithmetic logic unit is a set of combinational logic circuits that receive one or more inputs to generate a result. In at least one embodiment, the processor utilizes the arithmetic logic unit to perform mathematical operations such as addition, subtraction, or multiplication. In at least one embodiment, the arithmetic logic unit is also used to perform logical operations such as AND / OR or XOR. In at least one embodiment, the arithmetic logic unit is stateless and consists of physical switching elements such as semiconductor transistors, which are arranged in the form of logic gates. In at least one embodiment, the arithmetic logic unit may internally operate as a stateful logic circuit with a correlated clock. In at least one embodiment, the arithmetic logic unit may be constructed as an asynchronous logic circuit whose internal state is not stored in a correlated register set. In at least one embodiment, the processor utilizes the arithmetic logic unit to perform combinational operations on operands stored in one or more of its registers and generate an output result that can be stored by the processor in another register or a memory location.

[0116] In at least one embodiment, after the processor retrieves and processes an instruction, it provides one or more inputs or operands to the arithmetic logic unit (ALU), causing the ALU to generate a result at least partially based on the instruction code provided to its inputs. In at least one embodiment, the instruction code provided by the processor to the ALU is at least partially based on instructions executed by the processor. In at least one embodiment, combinational logic within the ALU processes the input signals and generates an output result, which is placed on a bus within the processor. In at least one embodiment, the processor selects a target register, memory location, output device, or output memory location on the output bus such that when the processor triggers a clock signal, the result generated by the ALU can be sent to that target location.

[0117] Therefore, in at least one embodiment, the computer system is configured to implement one or more services that individually or collectively perform the process operations described herein; and such a computer system is equipped with corresponding hardware and / or software to support the execution of these operations. Further, the computer system implementing at least one embodiment of this disclosure may be a single device; in another embodiment, it may also be a distributed computer system consisting of multiple devices operating in different modes, such that the distributed computer system performs the operations described herein, while a single device does not need to perform all the operations.

[0118] The use of any and all examples or exemplary expressions (such as "such as") provided herein is intended only to better illustrate embodiments of this disclosure and, unless otherwise stated, shall not limit the scope of this disclosure. Nothing in this specification should be construed as meaning that any unclaimed element is necessary for carrying out this disclosure.

[0119] In the specification and claims, the terms “coupled” and “connected” and their derivatives may be used. It should be understood that these terms are not intended to be synonyms. Rather, in certain examples, “connected” or “coupled” can be used to indicate that two or more elements are in direct or indirect physical or electrical contact with each other. “Coupled” can also refer to two or more elements that are not in direct contact but can still cooperate or interact with each other.

[0120] Unless otherwise expressly stated, it should be understood that the terms used throughout this specification (such as “processing,” “calculation,” “operation,” “determine,” etc.) refer to the actions and / or processes of a computer or computing system (or similar electronic computing device) that manipulate and / or convert data represented as physical quantities (such as electronic quantities) in the computing system registers and / or memory into similar data represented as physical quantities in the computing system memory, registers, or other such information storage, transmission, or display devices.

[0121] Similarly, the term "processor" can refer to any device or part of a device that retrieves electronic data from registers and / or memory and converts that electronic data into other electronic data that can be stored in registers and / or memory. As a non-limiting example, a "processor" can be a central processing unit (CPU) or a graphics processing unit (GPU). A "computing platform" can include one or more processors. As used herein, a "software process" can include, for example, software and / or hardware entities that perform operations over time, such as tasks, threads, and intelligent agents. Furthermore, each process can refer to multiple processes for executing instructions sequentially or in parallel, continuously or intermittently. In at least one embodiment, the terms "system" and "method" are used interchangeably herein, provided that a system can embody one or more methods, and a method can be considered a system.

[0122] This document may refer to acquiring, collecting, receiving, or inputting analog or digital data to a subsystem, computer system, or computer-implemented device. In at least one embodiment, the process of acquiring, collecting, receiving, or inputting analog and digital data can be implemented in various ways, such as receiving data as parameters of a function call or application programming interface (API) call. In at least one embodiment, the process of acquiring, collecting, receiving, or inputting analog or digital data can be implemented by transmitting data through a serial or parallel interface. In at least one embodiment, the process of acquiring, collecting, receiving, or inputting analog or digital data can be implemented by transmitting data from a provider entity to a recipient entity via a computer network. This document may also refer to providing, outputting, transmitting, sending, or presenting analog or digital data. In at least one embodiment, the process of providing, outputting, transmitting, sending, or presenting analog or digital data can be implemented by transmitting data as input or output parameters of a function call, parameters of an application programming interface (API), or parameters of an inter-process communication mechanism.

[0123] While the description herein illustrates exemplary implementations of the technology, other architectures may also be used to implement the functionality, and all such architectures are intended to fall within the scope of this disclosure. Furthermore, although specific allocation of responsibilities has been defined above for descriptive purposes, functions and responsibilities may be allocated and divided in different ways depending on the specific circumstances.

[0124] Furthermore, although the subject matter herein has been described in language specific to structural features and / or method steps, it should be understood that the subject matter defined by the appended claims is not necessarily limited to the specific features or steps described above. Rather, the specific features and steps described above are disclosed as exemplary forms for implementing the claims.

Claims

1. A system for leak detection in a computing environment, comprising: A shaped leak sensor, comprising an insulating material and a plurality of conductive traces, wherein the shaped leak sensor is configured to be shaped to at least match a layout around a component in the computing environment; and A detector is used to monitor input from the forming leak sensor to determine one or more states among a plurality of states associated with the forming leak sensor.

2. The system according to claim 1, wherein, The plurality of conductive traces are included on a first side of the insulating material, and wherein an adhesive is included on a second side of the insulating material.

3. The system according to claim 1, wherein, The shaped leak sensor is shaped to adapt to at least a portion of a central processing unit (CPU), graphics processing unit (GPU), or data processing unit (DPU), wherein the CPU, GPU, or DPU constitutes the component in the computing environment.

4. The system according to claim 1, wherein, The insulating material is a flexible insulating material and is one or more of polyimide, polyamide, polyester, polyethylene naphthalate or polytetrafluoroethylene, and wherein the plurality of conductive traces include one or more of gold, graphite, silver or nickel.

5. The system according to claim 1, further comprising: Foam or volumetric material around the shaped leak sensor is used to support the collection of leaking fluid from the computing environment.

6. The system according to claim 1, further comprising: One or more pins or sockets, which are used to allow the shaped leak sensor to be associated with another shaped leak sensor via a plug-in arrangement, and to be associated with the detector directly or through the other shaped leak sensor.

7. The system according to claim 1, further comprising: At least one pair of parallel conductive traces of the plurality of conductive traces, wherein the at least one pair of parallel conductive traces is configured to be shaped to provide the shaped leak sensor, and is configured to provide redundancy for sensing leaks when at least one of the at least one pair of parallel conductive traces is shaped.

8. The system according to claim 1, wherein, The shaped leak sensor is also configured to be calibrated or tested for one of the different applications in the computing environment, wherein at least the calibration is performed in a dry-state version of the computing environment to determine reference values ​​for the plurality of states, and wherein the detector is a voltage-based leak detection device configured to distinguish the plurality of states in part based on different voltages provided in the input relative to the reference values.

9. The system according to claim 8, wherein, At least the test is based in part on at least one electrical component coupled to one or more of the shaped leak sensors to simulate different resistances.

10. A molded leak sensor, comprising an insulating material, a plurality of conductive traces printed or applied thereon on a first side, and an adhesive on a second side, wherein, The shaped leak sensor is shaped by one or more cuts to at least match the layout around the component in the computing environment, and wherein the shaped leak sensor provides input to a detector to be able to monitor leaks and to be able to determine one or more of a plurality of states associated with the shaped leak sensor.

11. The molded leak sensor according to claim 10, wherein, The insulating material is a flexible insulating material, including one or more of polyimide, polyamide, polyester, polyethylene naphthalate, or polytetrafluoroethylene, and wherein the plurality of conductive traces include one or more of gold, graphite, silver, or nickel.

12. The molded leak sensor according to claim 10, further comprising: Foam or volumetric material around the shaped leak sensor to support the collection of leaking fluid from the computing environment.

13. The molded leak sensor according to claim 10, further comprising: One or more pins or sockets, which are used to allow the shaped leak sensor to be associated with another shaped leak sensor via a plug-in arrangement, and to be associated with the detector directly or through the other shaped leak sensor.

14. The molded leak sensor according to claim 10, further comprising: At least one pair of parallel conductive traces of the plurality of conductive traces, wherein the at least one pair of parallel conductive traces is configured to be shaped to provide the shaped leak sensor, and is configured to provide redundancy for sensing leaks when at least one of the at least one pair of parallel conductive traces is shaped.

15. The molded leak sensor according to claim 10, wherein, The shaped leak sensor is also configured to be calibrated or tested for one of the different applications in the computing environment, wherein at least the calibration is performed in a dry-state version of the computing environment to determine reference values ​​for the plurality of states, and wherein the detector is a voltage-based leak detector and is configured to distinguish the plurality of states in part based on different voltages provided in the input relative to the reference values.

16. The molded leak sensor according to claim 15, wherein, At least the test is based in part on at least one electrical component coupled to one or more of the shaped leak sensors to simulate different resistances.

17. A method for leak detection in a computing environment, comprising: A molded leak sensor is fabricated, the molded leak sensor comprising an insulating material, a plurality of conductive traces printed or applied thereon on a first side, and an adhesive on a second side; The shape in the shaped leak sensor should at least match the layout around the component in the computing environment; as well as The detector is used to monitor the input from the forming leak sensor to determine one or more of a plurality of states associated with the forming leak sensor.

18. The method according to claim 17, wherein, The insulating material is a flexible insulating material, including one or more of polyimide, polyamide, polyester, polyethylene naphthalate, or polytetrafluoroethylene, and wherein the plurality of conductive traces include one or more of gold, graphite, silver, or nickel.

19. The method of claim 17, further comprising: The foam or volumetric material around the shaped leak sensor is designed to support the collection of leaking fluid from the computing environment.

20. The method of claim 17, further comprising: The shaped leak sensor can be associated with another shaped leak sensor using a plug-in arrangement of one or more pins or sockets; and The detector may be associated directly or through the other shaped leak sensor.

21. The method of claim 17, further comprising: Enable at least one pair of parallel conductive traces among the plurality of conductive traces; as well as The at least one pair of parallel conductive traces are configured to be shaped to provide the shaped leak sensor, and redundancy is provided when at least one of the at least one pair of parallel conductive traces is shaped to sense a leak.

22. The method of claim 17, further comprising: The shaped leak sensor is calibrated or tested for one of the different applications in the computing environment, wherein at least the calibration is performed in a dry-state version of the computing environment to determine reference values ​​for the plurality of states, and wherein the detector is a voltage-based leak detector configured to distinguish the plurality of states in part based on different voltages provided in the input relative to the reference values.

23. A data center, comprising: One or more racks, wherein the one or more racks include one or more server trays; One or more components of the one or more racks, the one or more components being used to perform at least a portion of the workloads in the data center; A shaped leak sensor, the shaped leak sensor comprising an insulating material and a plurality of conductive traces, wherein the shaped leak sensor is configured to be shaped to match at least the layout around one or more components; as well as A detector is used to monitor input from the forming leak sensor to determine one or more states among a plurality of states associated with the forming leak sensor.

Images