Organic rankine cycle for data center electronics cooling and thermal energy recovery

Abstract

A technique for recovering power from data center waste heat involves an organic Rankine cycle that includes pumping an organic working fluid through a low-temperature preheater chamber configured to exchange relatively low-temperature waste heat from low-temperature data storage devices to the organic working fluid flowing through the preheater chamber, pumping the working fluid through a high-temperature evaporator thermally coupled to the preheater chamber and configured to exchange relatively high-temperature waste heat from high-temperature CPUs to the working fluid flowing through the evaporator, and pumping working fluid vapor from the evaporator to an expander to generate power via expansion of the working fluid. The preheater and evaporator can be incorporated directly with a housing cabinet that houses various heat-generating components.

Description

Cross-references to related applications

[0001] This application claims the benefit and priority of U.S. non-provisional patent application serial number 18 / 412,510, filed on January 13, 2024. Technical Field

[0002] The embodiments of the present invention relate generally to electronic equipment, and more specifically, to a method for recovering power from waste heat in data centers. Background Technology

[0003] As the number and functionality of networked computing systems grow, the demand for storage system capacity also increases. Cloud computing and large-scale data processing further increase the need for digital data storage systems capable of transmitting and holding massive amounts of data. One way to provide sufficient data storage for data centers is to use data storage device arrays. Many data storage devices can be housed in electronic enclosures (sometimes called "racks"), which are typically modular units that hold and operate individual data storage devices, computer processors, switches, routers, and other electronic equipment (collectively called "boxes"). Data centers typically consist of many rack-mountable "boxes" for storing and processing large amounts of data. Data storage and data processing are known to consume significant amounts of power. This power is also known to dissipate as heat, requiring substantial cooling and associated costs.

[0004] Any method that may be described in this section is a feasible method, but not necessarily a method that has been previously conceived or implemented. Therefore, unless otherwise stated, no method described in this section should be considered prior art simply because it is included in this section. Attached Figure Description

[0005] The embodiments are illustrated in the accompanying drawings by way of example rather than limitation, in which the same reference numerals refer to similar elements and wherein:

[0006] Figure 1A This is a plan view illustrating a hard disk drive (HDD) according to one implementation scheme;

[0007] Figure 1B This is a block diagram illustrating a solid-state drive (SSD) according to one implementation scheme;

[0008] Figure 2 This is a perspective view illustrating a data center;

[0009] Figure 3 This is a block diagram illustrating a data storage system architecture based on one implementation scheme;

[0010] Figure 4This is a functional diagram illustrating a data center organic Rankine loop architecture according to one implementation scheme;

[0011] Figure 5 This is a diagram illustrating a heat recovery data storage system according to one implementation scheme; and

[0012] Figure 6 This is a flowchart illustrating a method for recovering power from waste heat in a data center according to one implementation scheme. Detailed Implementation

[0013] A method for recovering power from waste heat in a data center is described. In the following description, numerous specific details are set forth for purposes of explanation in order to provide a thorough understanding of the embodiments of the invention described herein. However, it will be apparent, however, that the embodiments of the invention described herein can be practiced without these specific details. In other instances, well-known structures and apparatuses are shown in block diagram form to avoid unnecessarily obscuring the embodiments of the invention described herein.

[0014] introduction

[0015] the term

[0016] References to phrases such as "an embodiment" and "one embodiment" herein are intended to mean that a particular feature, structure, or characteristic described is included in at least one embodiment of the invention. However, instances of such phrases do not necessarily refer to the same embodiment.

[0017] As used herein, the term "substantially" should be understood to describe features that are mostly or nearly structured, configured, sized, etc., but in practice, manufacturing tolerances and other factors may cause structures, configurations, dimensions, etc., to not always or necessarily be as precise as described. For example, describing a structure as "substantially vertical" would give the term its general meaning, implying that the sidewalls are vertical for all practical purposes, but may not be precisely at 90 degrees throughout.

[0018] While terms such as “optimal,” “minimum,” “maximum,” “maximize” may not have certain values ​​associated with them, if used herein, it is intended that those skilled in the art will understand that such terms will encompass values, parameters, measures, etc., that influence in a beneficial direction consistent with the whole of this disclosure. For example, describing the value of something as “minimum” does not require that the value is actually equal to some theoretical minimum (e.g., zero), but should be understood in a practical sense as the corresponding objective being to move that value toward the theoretical minimum in a beneficial direction.

[0019] Physical description of exemplary operating scenarios - data storage systems and data centers

[0020] There is a commercial demand for high-capacity digital data storage and processing systems (typically data storage systems or "DSS") where multiple data storage devices (DSDs), such as hard disk drives (HDDs), solid-state drives (SSDs), tape drives, hybrid drives, etc., are housed in a common enclosure. Data storage systems (often referred to as "servers") typically comprise a large enclosure (or "chamber") housing multiple slots or tracks in which rows of DSDs are mounted, whereby each chamber can then be placed or slid into a corresponding shelf or track within a rack or cabinet. Each DSD is communicatively coupled to a system controller, such as via a backplane or otherwise, whereby the system controller may be housed, for example, within the DSS enclosure to control these DSDs. Additionally, the system controller may be housed elsewhere within the rack for broader control of the storage / computing system. Typically, each rack may also house routers, switches, patch panels, storage servers, application servers, power supplies, cooling fans, etc.

[0021] Typically, a data center (or more generally, a "massive storage system") can be likened to an extreme version of a data storage system (or multiple data storage systems working together), including the power, cooling, space, and corresponding network infrastructure (e.g., routers, switches, firewalls, application delivery controllers, etc.) required to store, manage, and share operational data. Expanding on this concept, a "hyperscale" data center generally refers to a facility that provides robust, scalable application and storage services to individuals or other businesses. Exemplary implementations of hyperscale computing include cloud and big data storage, web services and social media platforms, and enterprise data centers, which can consist of thousands of servers linked via ultra-high-speed fiber optic networks.

[0022] Figure 2This is an illustrative perspective view of a data center. For illustrative purposes only, data center 200 includes multiple racks 201a, 201b-201n positioned along rows, where n represents any number of racks that can vary depending on the specific implementation. Each rack 201a-201n houses multiple DSS servers 202a, 202b-202m (or simply "DSS 202a-202m"), where m represents any number of DSS enclosures that can vary depending on the specific implementation. Each DSS 202a-202m typically includes multiple DSDs (e.g., HDDs and / or SSDs), as discussed elsewhere in this document. It is well known and readily understood that the various electronic components constituting a data center such as data center 200 naturally generate a non-small (in fact, a large) amount of heat (i.e., primarily due to resistance). Therefore, how to dissipate the heat generated by the operation of electronic components, especially the large amount of heat generated by a large data center, remains an ongoing challenge.

[0023] The example data storage system may include multiple DSDs, such as SSDs and / or HDDs, each of which communicates with and is controlled by the system controller (or, for example, an I / O (input / output) controller or an I / O computing board) via a communication interface circuit system according to a corresponding communication protocol. Figure 3 This is a block diagram illustrating a data storage system architecture according to one implementation. Example architecture 300 illustrates a data storage system 302 including multiple data storage devices (DSDs) 304a (DSD1), 304b (DSD2), and 304n (DSDn), where n represents any number of DSDs (e.g., SSDs and / or HDDs) that may vary depending on the specific implementation. Each DSD 304a-304n communicates with and is controlled to some extent by a storage system controller 312 via a communication interface 322 (e.g., an electronic circuitry system including electrical connections) according to a corresponding communication protocol 323. Each DSD 304a-304n includes a corresponding non-volatile memory (NVM) 306 (e.g., typically in the form of electronic non-volatile memory, such as in the case of an SSD) controlled by a corresponding DSD controller 308 (“DSD CNTLR” 308). Figure 1A Non-volatile memory components 170a-170n, or, in the case of HDDs, rotating disk media, such as Figure 1BThe recording medium 120). Each DSD controller 308 includes at least a memory 309 and a processor 311, while the system controller 312 also includes at least a memory 313 and a processor 315. The DSD controller 308 and the system controller 312 may each be implemented in any form and / or combination of software, hardware, and firmware. The electronic controller in this context typically includes a circuit system such as one or more processors for executing instructions, and may be implemented as a system-on-a-chip (SoC) electronic circuit system, which, for non-limiting examples, may include memory, a microcontroller, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), hardwired logic, analog circuit systems, and / or combinations thereof.

[0024] Data storage system 302 may be communicatively coupled to host 350, which may be implemented as a hardware machine on which executable code is executed (for a non-limiting example, a computer or hardware server, etc.), or as software instructions that can be executed by one or more processors (for a non-limiting example, a software server such as a database server, application server, media server, etc.). Host 350 typically refers to a client of data storage system 302 and has the ability to make read and write requests (“I / O”) to data storage system 302. Note that system controller 312 may also be referred to as a “host”, as the term is generally used to refer to any device that makes I / O calls to data storage devices or device arrays, such as DSD 304a-304n. For a non-limiting example, host 350 interacts with one or more DSDs 304a-304n via interface 322 (e.g., physical and electrical I / O interfaces) for transferring data to and from DSDs 304a-304n, such as via a bus or network such as Ethernet or Wi-Fi or a bus standard such as Serial Advanced Technology Attachment (SATA), Peripheral Component Interconnect (PCI) High Speed ​​(PCIe), Small Computer System Interface (SCSI), or Serial Attached SCSI (SAS).

[0025] Context

[0026] To recap, data centers consume vast amounts of power during operation, generating significant amounts of heat, which requires substantial cooling. Furthermore, the extensive cooling required for data centers incurs high monetary costs and often leads to substantial environmental costs. One possible approach to addressing this challenge could be to utilize compact thermoelectric generators that directly convert heat flux into electrical energy. However, thermoelectric generators are considered relatively inefficient. Another possible approach could be to utilize a simple Stirling cycle for self-cooling. However, this approach is also considered relatively inefficient due to its low compression ratio and low flux (i.e., the working fluid is a gas), and is therefore not considered suitable for high-power scenarios such as data centers.

[0027] Organic Rankine Cycle (ORC) for Data Centers

[0028] Compared to the methods mentioned, the use of the Organic Rankine Cycle (“ORC”) for waste heat recovery / power generation in a data center scenario is considered suitable for the intended purpose, at least in part because of its high flux (i.e., utilization of boiling and condensation), relatively high efficiency of the thermodynamic cycle, and the unique saturation profile of the organic working fluid. To elaborate on the use of the organic working fluid in the ORC, the nature of water's saturation profile limits its benefits for the intended purpose, as the expansion of saturated steam will enter the two-phase region, and liquid water reduces efficiency and significantly shortens turbine life. Furthermore, superheating is challenging due to the low flux of the heating steam. In contrast, the organic working fluid has a unique saturation profile where the expanding saturated fluid further generates steam in the superheated region, allowing the turbine to operate efficiently and utilize flux more aggressively, such as by utilizing expansion beyond the saturation boundary. In view of the foregoing, and according to one implementation, ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) designation R365 refrigerant (C4H5F5, chemical name 1,1,1,3,3-pentafluorobutane) is considered a suitable organic working fluid for the intended purpose, at least in part due to maximum Carnot efficiency and ideal cycle efficiency, and further taking into account the associated temperature difference and reasonable pressure range.

[0029] Figure 4This is a functional diagram illustrating a data center organic Rankine cycle architecture according to one embodiment. Here, according to one embodiment, the hot section of ORC 400 (depicted with less dense crosshairs to indicate the "red" or "hot" section) is referred to as multi-stage because it utilizes a preheater 402 (also referred to as a feed working fluid heater) combined with an evaporator 404 (also referred to as a "boiler" or "superheater"). Here, the term "evaporator" is intended to include evaporation and / or boiling and / or superheating functions, thereby achieving a subcritical or supercritical cycle, and these terms are used interchangeably herein. Thus, some cycles can undergo a supercritical phase, thereby operating at sufficiently high pressures such that when the working fluid is heated in the evaporator (boiler / superheater) 404, the fluid properties exceed their saturation curves, and there is no significant phase change in the working fluid, i.e., where the fluid no longer truly "evaporates". Typically, for a Rankine thermodynamic cycle, the working fluid is pumped by pump 410 or otherwise passed through a preheater 402 (also referred to as "preheater chamber 402") and an evaporator 404, where condensed working fluid is supplied to pump 410 from a condenser 408 (depicted with denser crosshairs to indicate the "blue" or "cold" section). Furthermore, the working fluid in vapor form exiting the evaporator 404 flows to an expander / generator 406 (or simply "expander 406," which may also be referred to as a turbine or turbine / generator) for expansion and power generation. As depicted, ORC 400 represents a closed thermodynamic cycle. While not strictly necessary, a closed cycle is at least partially more efficient because the working fluid can expand to the saturation pressure at atmospheric temperature. For a non-limiting example, the R365 working fluid can expand to 0.56 atm (atmospheres) at 25°C, whereas in an open cycle, the working fluid would expand to 1 atm (atmospheres).

[0030] Here, the multi-stage thermal cycle preheater 402 includes a low-temperature preheater chamber configured to receive heat from one or more relatively low-temperature electronic components (e.g., from a data center 200). Figure 2 The data center absorbs low-temperature waste heat and exchanges it with an organic working fluid flowing through preheater 402. According to one embodiment, the low-temperature waste heat is drawn from relatively low-temperature IT (information technology) electronic components (such as data storage devices, e.g., HDDs and SSDs, etc.). Figure 3 The data storage system 302's DSD1-DSDn (304a-304n), for example, dissipates heat at approximately 60°-80°C. The relatively low-temperature IT electronic components can be housed within the electronic device housing (e.g., see...). Figure 5The housing 501). Furthermore, the multi-stage thermal cycle evaporator 404 includes a high-temperature evaporator in series with the preheater 402 in fluid communication, and is configured to absorb high-temperature waste heat from one or more relatively high-temperature electronic components and exchange the high-temperature waste heat to an organic working fluid flowing through the evaporator 404. According to one embodiment, the high-temperature waste heat is drawn from relatively high-temperature IT electronic components (such as a system CPU, e.g., Figure 3 The processor 315 of the system controller 312 of the data storage system 302, for example, dissipates heat of about 100°C. The relatively high-temperature IT electronic components can also be housed in the same or different electronic device housings (e.g., see...). Figure 5 The ORC 400 is housed in a casing 501. The ORC 400 also includes an expander 406, to which heated and evaporated organic working fluid is transferred from the evaporator 404 to expand within the expander 406, thereby extracting power from waste heat in the data center.

[0031] because Figure 4 The ORC 400 is a functional diagram and is therefore not intended to strictly convey a single possible or preferred architecture where only a single circulating component (i.e., a preheater 402, an evaporator 404, an expander 406, a condenser 408, and a pump 410) is implemented for the entire data center. Instead, the granularity of the ORC 400 component architecture can vary depending on the specific implementation based on (for non-limiting examples) the desired thermodynamic cycle efficiency and the associated working fluid, as well as the corresponding saturation curves, flow rates, pressures, etc., and practical and logical considerations. For example, the ORC design can be implemented at the data center level, whereby the working fluid is pumped, piped, and routed to expand in expander 406; then pumped, piped, routed, and heat-exchanged with all high-temperature components (in series or parallel) throughout the data center; then pumped, piped, routed, and thus heat-exchanged with all low-temperature components (in series or parallel) throughout the data center. In another example, the ORC design can be implemented at the rack level, whereby the working fluid is combined and pumped, piped, and routed to expand in expander 406; then pumped, piped, routed, and heat-exchanged with all the high-temperature components (in series or in parallel) in the rack; then pumped, piped, routed, and thus heat-exchanged with all the low-temperature components (in series or in parallel) in a given rack.

[0032] Furthermore, according to one embodiment, the preheater 402 and evaporator 404 are thermally coupled to an electronic device housing (e.g., a "box") containing relatively low-temperature and relatively high-temperature electronic components, and an organic working fluid flows from the preheater 402 to the evaporator 404. In other words, at least in part to maintain line replaceable unit (LRU) capability, the preheater 402 and evaporator 404, together with the storage / computing electronic components that generate waste heat, are configured as part of a given box or storage / computing unit.

[0033] Figure 5 This is a diagram illustrating a heat recovery data storage system according to one embodiment. The data storage system 500 (or "DSS 500") is depicted as an LRU or enclosure, such as its commercial form, where it is integrated or otherwise incorporated into various levels of an ORC. The DSS 500 is depicted in a simplified schematic form and includes components configured similarly to preheater 402 ( Figure 4 The preheater chamber 502, which functions as a preheater, and the evaporator 404, which is configured to function similarly to the evaporator. Figure 4 The preheater chamber 502 and the evaporator 504 are both thermally coupled to a portion of the housing 501 (e.g., in a configuration / location where heat exchange is possible). Here, the housing 501 to which the preheater chamber 502 and the evaporator 504 are thermally (and mechanically coupled) is the same housing that also houses relatively low-temperature electronic components 512 (also referred to as "low-temperature component 512") and relatively high-temperature electronic components 514 (also referred to as "high-temperature component 514"). As discussed and according to one embodiment, for a non-limiting example, the low-temperature component 512 may include a data storage device, such as a hard disk drive (HDD) and / or a solid-state drive (SDD) (see, for example...). Figure 3 The data storage system 302 (DSD1-DSDn (304a-304n)) and any other similar low-temperature components that dissipate heat of approximately 60°-80°C, and for a non-limiting example, the high-temperature component 514 may include a computing processor / central processing unit (CPU) (see, for example, see...). Figure 3 The data storage system 302 includes the system controller 312 (processor 315) and any other high-temperature components that emit heat of approximately 100°C.

[0034] Operationally, "cold" organic working fluid 511 enters and flows through preheater chamber 502, thereby cooling cryogenic component 512 via the exchange of corresponding waste heat from cryogenic component 512 to cold working fluid 511. Continuing, "warm" organic working fluid 513 flows from preheater chamber 502 to evaporator 504, thereby cooling cryogenic component 514 via the exchange of corresponding waste heat from cryogenic component 514 to warm working fluid 513 from preheater chamber 502. Now, "hot" working fluid 515 flows from evaporator 504 in its desired phase to one or more expanders, such as expander 406 (…). Figure 4 The desired configuration is suited to its intended purpose of generating electricity through expansion. It is conceivable that each DSS 500 may be configured with its own expander 406. According to one embodiment, to aid heat exchange, the preheater chamber 502 is thermally coupled to the electronics housing 501 via one or more first heat exchangers 503, and the evaporator 504 is thermally coupled to the electronics housing 501 via one or more second heat exchangers 505.

[0035] Since the given DSS 500 chamber can also include additional electronic components that can dissipate heat in other temperature ranges away from the temperature ranges of the cryogenic component 512 and the high-temperature component 514, a system such as the DSS 500 can be realized, in which one or more, or even lower, cryogenic preheater chambers 502a, 502b are in fluid communication with cryogenic preheater chamber 502 and are configured to absorb waste heat at a lower temperature below the cryogenic waste heat from one or more relatively cryogenic electronic components 512a, 512b, and exchange the lower-temperature waste heat to the organic working fluid flowing through each lower-temperature preheater chamber 502a, 502b to reach cryogenic preheater chamber 502. Thus, a gradient of preheating function can be realized to gradually preheat the working fluid for transfer to evaporator 504, thereby further heating, boiling, and evaporating. Here, the lower-temperature electronic components 512a, 512b can also be thermally coupled to the electronic device housing 501 via one or more corresponding heat exchangers 503a, 503b to assist the heat exchange process.

[0036] Modular configurations, such as the ORC 400, are achieved using electronic units such as the DSS 500. Figure 4The data center waste heat recovery ORC can also be modular or distributed to the desired extent. According to one embodiment, each of the multiple preheater chambers 502 is thermally coupled to a corresponding electronics housing 501, with a corresponding relatively low-temperature electronics component 512 housed within the corresponding electronics housing. Similarly, each of the multiple evaporators 504 is thermally coupled to a corresponding electronics housing 501, with a corresponding relatively high-temperature electronics component 514 housed within the corresponding electronics housing, and the corresponding preheater chamber 502 is thermally coupled to its corresponding electronics housing. Therefore, a manifold can be configured to fluidly connect two or more evaporators in the evaporators 504 before being delivered to the expander 406. For a non-limiting example, the hot working fluid 515 (i.e., steam) leaving each evaporator 504 can be combined at the rack level en route to a shared expander 406 for multiple racks, or each rack can be configured with its own expander 406, etc.

[0037] Although the heat exchange function between the low-temperature component 512 and the preheater chamber 502 and between the high-temperature component 514 and the evaporator 504 is referenced Figure 5 Described as utilizing surface contact between corresponding components, but further envisioning a two-phase immersion cooling technique to facilitate such heat exchange as an alternative to or enhancement of surface contact exchange. Typically, in two-phase immersion cooling, electronic components are directly immersed in a dielectric liquid, where heat from the electronic components causes the liquid to boil, thereby generating vapor rising from the liquid. Therefore, and according to one embodiment, to maintain efficiency, cryogenic electronic components (such as cryogenic component 512) are immersed in a preheater chamber (such as preheater 402). Figure 4 The organic working fluid in the evaporator (such as evaporator 404) is immersed in the high-temperature electronic component 514. Figure 4 The low-temperature component 512 may be housed in a corresponding sealed housing (e.g., sealed housing 501) immersed in the working fluid corresponding to the preheater 402 (e.g., cold working fluid 511 to warm working fluid 513), while the high-temperature component 514 may be housed in a corresponding sealed housing (e.g., sealed housing 501) immersed in the working fluid corresponding to the evaporator 404 (e.g., warm working fluid 513 to hot working fluid 515).

[0038] This arrangement can be developed and implemented at the rack level, where a liquid-sealed rack houses a liquid-sealed housing (accommodating cryogenic component 512 and / or high-temperature component 514) within the working fluid. If each tank (such as DSS 500) contains a mixture of cryogenic component 512 and high-temperature component 514, the boiling temperature and pressure are determined by the lowest-temperature component, and therefore the cycle will have a relatively low regeneration efficiency. In this scenario, preheater chambers 402, 502 and evaporators 404, 504 are effectively configured as integrated chambers where cryogenic electronic component 512 and high-temperature electronic component 514 are immersed in the organic working fluid. Thus, and according to one embodiment, the tanks can be connected in parallel, and all steam will be collected to enter expander 406. If each tank is dedicated to a single type of component (e.g., cryogenic or high-temperature), the tanks can be connected in series, where the cryogenic component is first in the line, and the feed working fluid gradually flows towards the tank with the higher-temperature component, eventually reaching the evaporator 404 corresponding to one or more tanks with the highest-temperature component. Here, steam generated from one rack can also be combined with steam generated from other racks before entering expander 406.

[0039] Alternatively, in an immersion cooling and regeneration scenario, the working fluid can be pumped through cold plates attached to the cryogenic component 512 and the high-temperature component 514, thereby preventing the working fluid from directly contacting the electronic components.

[0040] Methods for recovering power from data centers

[0041] Figure 6 This is a flowchart illustrating a method for recovering power from waste heat in a data center according to one implementation scheme. Applicable data centers can be such as... Figure 2 The 200 data centers exemplified here house multiple data storage systems, such as Figure 3 The data storage system 302, as illustrated, can be configured as a container such as the heat recovery DSS 500.

[0042] At frame 602, an organic working fluid is pumped through a cryogenic preheater chamber configured to exchange cryogenic waste heat from one or more relatively cryogenic electronic components with the organic working fluid flowing through the preheater chamber. For example, "cold" organic working fluid 511 ( Figure 5 For example, via pump 410 ( Figure 4 Pumped through cryogenic preheater chamber 402 ( Figure 4 ), 502 Figure 5 The cryogenic preheater chamber is configured to receive heat from one or more relatively cryogenic electronic components 512 ( Figure 5Low-temperature waste heat (such as HDD and / or SSD) is exchanged to organic working fluids 511-513 flowing through preheater chambers 402 and 502.

[0043] As discussed elsewhere in this document, before moving the working fluid through the low-temperature preheater chambers 402, 502, the organic working fluid 511 can be drawn from one or more lower-temperature preheater chambers 502a, 502b ( Figure 5 Pumping, one or more lower temperature preheater chambers are thermally coupled to low temperature preheater chamber 502 and configured to pump lower temperature waste heat having a lower temperature than low temperature waste heat from one or more relatively low temperature electronic components 512a, 512b ( Figure 5 The working fluid is exchanged with the organic working fluid flowing through the lower-temperature preheater chambers 502a and 502b. Similarly, after moving the working fluid through the low-temperature preheater chambers 402 and 502 and after moving the working fluid through the evaporator 404 ( Figure 4 ), 504 Figure 5 Previously, the organic working fluid 513 could be pumped from the low-temperature preheater chambers 402 and 502 to another thermally coupled preheater chamber, which was configured to exchange intermediate-temperature waste heat with a temperature higher than that of the low-temperature waste heat but lower than that of the high-temperature waste heat from one or more relatively intermediate-temperature electronic components to the organic working fluid 513 flowing from the low-temperature preheater chambers 402 and 502.

[0044] At frame 604, an organic working fluid is pumped through a high-temperature evaporator, which is thermally coupled to a preheater chamber and configured to exchange high-temperature waste heat from one or more relatively high-temperature electronic components with the organic working fluid flowing through the evaporator. For example, a "warm" organic working fluid 513 ( Figure 5 The fluid is pumped (e.g., via pump 410) through high-temperature evaporators 404, 504, which are thermally coupled to preheater chambers 402, 502 and configured to pump fluid from one or more relatively high-temperature electronic components 514. Figure 5 High-temperature waste heat (such as that from a CPU) is exchanged to organic working fluid 513-515 flowing through evaporators 404 and 504. Alternatively, the immersion cooling / regeneration techniques described elsewhere in this document can be implemented at execution blocks 602 and 604.

[0045] At box 606, heated organic working fluid is pumped from the evaporator to the expander to generate electricity (or generally perform work) via the expansion of the organic working fluid. For example, "hot" organic working fluid 515 is pumped from evaporators 404, 504 (e.g., via pump 410) to expander / generator 406. Figure 4 ), used to generate electricity (or generally to do work) via the expansion of an organic working fluid (i.e., now at least partially in the gas phase) 515.

[0046] In light of the foregoing, this paper describes a method for recovering / regenerating power from waste heat in data centers via an organic Rankine cycle. Therefore, such methods can enable the elimination of cooling fans and associated additional power consumption, operating and environmental costs, noise, vibration, etc., from data centers and / or component systems and equipment.

[0047] Hard drive configuration

[0048] As discussed, the implementation scheme can be used in a data center context employing multiple data storage devices (DSDs) such as hard disk drives (HDDs). Therefore, according to one implementation scheme, Figure 1A A floor plan of an illustrative HDD 100 is shown to illustrate exemplary operating components, at least in part illustrating associated operating waste heat sources within a data center.

[0049] Figure 1A The functional arrangement of components of an HDD 100, including a slider 110b, is illustrated. The slider includes a magnetic read / write head 110a. The slider 110b and head 110a may be collectively referred to as the head slider. The HDD 100 includes at least one head gimbal assembly (HGA) 110 with the head slider, a lead suspension 110c typically attached to the head slider via a bend, and a load beam 110d attached to the lead suspension 110c. The HDD 100 also includes at least one recording medium 120 rotatably mounted on a spindle 124 and a drive motor (not visible) attached to the spindle 124 for rotating the medium 120. The read / write head 110a (also referred to as a transducer) includes a write element and a read element for writing and reading information stored on the medium 120 of the HDD 100, respectively. The medium 120 or multiple disk media can be attached to the spindle 124 using a disk clip 128.

[0050] HDD 100 also includes an arm 132 attached to HGA 110, a carriage 134, and a voice coil motor (VCM) including an armature 136 and a stator 144. The armature includes a voice coil 140 attached to the carriage 134, and the rotor includes a voice coil magnet (not visible). The armature 136 of the VCM is attached to the carriage 134 and configured to move the arm 132 and HGA 110 to access portions of the media 120. They are collectively mounted on a pivot shaft 148 having an inserted pivot bearing assembly 152. In the case of an HDD with multiple disks, the carriage 134 may be referred to as an "E-block" or comb because the carriage is arranged to carry a linked array of arms, thus giving it a comb-like appearance.

[0051] An assembly including a head universal joint assembly (e.g., HGA 110) with a bend to which the head slider is coupled, an actuator arm (e.g., arm 132) and / or a load beam to which the bend is coupled, and an actuator (e.g., VCM) to which the actuator arm is coupled, can be collectively referred to as a head stack assembly (HSA). However, an HSA may include more or fewer components than those described. For example, an HSA may refer to an assembly that also includes electrical interconnect components. Generally, an HSA is an assembly configured to move the head slider to access a portion of the medium 120 for read and write operations.

[0052] Referring further to Figure 1, electrical signals, including write signals to and read signals from the head 110a (e.g., current to the voice coil 140 of the VCM), are transmitted by a flexible cable assembly (FCA) 156 (or “flexible cable”). The interconnect between the flexible cable 156 and the head 110a may include an arm electronics (AE) module 160, which may have an onboard preamplifier for the read signal and other read and write channel electronics. The AE module 160 may be attached to a carriage 134, as shown. The flexible cable 156 may be coupled to an electrical connector block 164, which in some configurations provides electrical communication via an electrical feedthrough provided by the HDD housing 168. The HDD housing 168 (or “housing base” or “substrate” or simply “base”) together with the HDD cover provides a semi-sealed (or hermetically sealed, in some configurations) protective enclosure for the information storage components of the HDD 100.

[0053] Other electronic components (including the disk controller and servo electronics including a digital signal processor (DSP)) provide electrical signals to the drive motor, the voice coil 140 of the VCM, and the magnetic head 110a of the HGA 110. The electrical signals provided to the drive motor cause it to rotate, thereby providing torque to the spindle 124, which is then transmitted to the medium 120 attached to the spindle 124. The medium 120 thus rotates in direction 172. The rotating medium 120 forms an air cushion that acts as an air bearing on which the air bearing surface (ABS) of the slider 110b rests, allowing the slider 110b to fly above the surface of the medium 120 without contacting the thin magnetic recording layer on which information is recorded. Similarly, in HDDs utilizing gases lighter than air (such as helium used in a non-limiting example), the rotating medium 120 forms an air cushion that acts as a gas or fluid bearing on which the slider 110b rests.

[0054] The electrical signal supplied to the voice coil 140 of the VCM enables the head 110a of the HGA 110 to access the track 176 on which information is recorded. Therefore, the armature 136 of the VCM swings through an arc 180, allowing the head 110a of the HGA 110 to access the individual tracks on the medium 120. Information is stored in multiple radially nested tracks on the medium 120, which are arranged in sectors (such as sector 184) on the medium 120. Accordingly, each track is composed of multiple sectorized track portions (or “track sectors”) such as sectorized track portions 188. Each sectorized track portion 188 may include recorded information and a data header containing error correction code information and a servo burst signal pattern, such as the ABCD-servo burst signal pattern (which is information identifying track 176). When accessing track 176, the read element of the head 110a of the HGA 110 reads a servo burst signal pattern, which provides a positioning error signal (PES) to the servo electronics. This controls the electrical signal supplied to the voice coil 140 of the VCM, enabling the head 110a to follow track 176. Upon locating track 176 and identifying a specific sectorized track portion 188, the head 110a either reads information from track 176 or writes information to track 176 according to instructions received by the disk controller from an external agent (such as the microprocessor of a computer system).

[0055] The electronic architecture of an HDD includes multiple electronic components for performing their respective HDD operating functions, such as a hard disk controller (“HDC”), an interface controller, an arm electronics module, a data channel, a motor driver, a servo processor, a buffer memory, etc. Two or more of these components may be combined on a single integrated circuit board called a “system-on-a-chip” (“SOC”). Some (if not all) of these electronic components are typically arranged on a printed circuit board coupled to the bottom side of the HDD, such as to the HDD housing 168.

[0056] This document refers to hard disk drives (such as the HDD 100 illustrated and described with reference to Figure 1) that may include information storage devices sometimes referred to as “hybrid drives.” A hybrid drive generally refers to a storage device that combines the functionality of a conventional HDD (see, for example, HDD 100) with a solid-state storage device (SSD) that uses non-volatile memory (such as flash memory or other solid-state (e.g., integrated circuit) memory) that is electrically erasable and programmable. Because the operation, management, and control of different types of storage media are typically different, the solid-state portion of a hybrid drive may include its own corresponding controller functionality, which may be integrated with the HDD functionality into a single controller. Hybrid drives can be built and configured to operate and utilize the solid-state portion in a variety of ways, such as, by way of non-limiting example, using the solid-state memory as cache memory for storing frequently accessed data, for storing I / O-intensive data, etc. Additionally, hybrid drives can be built and configured essentially as two storage devices, namely a conventional HDD and an SSD, in a single housing, with one or more interfaces for host connectivity.

[0057] Solid State Drive Configuration

[0058] As discussed, the implementation scheme can be used in data center scenarios employing multiple data storage devices (DSDs) such as solid-state drives (SSDs). Therefore, Figure 1B This is a block diagram illustrating an example operating scenario in which embodiments of the present invention can be implemented, at least partially illustrating associated operational waste heat sources within a data center. Figure 1B An example of a general-purpose SSD architecture 150 is illustrated, in which SSD 152 is communicatively coupled to host 154 via main communication interface 156. Implementation schemes are not limited to this. Figure 1B The described configuration, on the contrary, can be utilized by implementing a system other than... Figure 1B This can be achieved using SSD configurations other than those illustrated. For example, the implementation could be implemented to operate in other environments that rely on non-volatile memory storage components for writing and reading data.

[0059] Host 154 broadly refers to any type of computing hardware, software, or firmware (or any combination thereof) that makes data I / O requests or calls to one or more memory devices. For example, host 154 can be an operating system running on a computer, tablet, mobile phone, or any type of computing device that typically contains or interacts with memory, such as host 350. Figure 3 The main interface 156 that couples the host 154 to the SSD 152 can be, for example, an internal bus of the storage system, a communication cable, or a wireless communication link.

[0060] Figure 1BThe illustrated example SSD 152 includes an interface 160, a controller 162 (e.g., a controller with firmware logic), an addressing function block 164, a data buffer cache 166, and one or more non-volatile memory components 170a, 170b-170n.

[0061] Interface 160 is the interaction point between component 152 (SSD) and host 154 in this context, and applies at both the hardware and software levels. This enables the component to communicate with other components via input / output (I / O) systems and associated protocols. Hardware interfaces are typically described by mechanical, electrical, and logical signals at the interface, as well as the protocols used to sequence them. Some non-limiting examples of general-purpose and standard interfaces include SCSI (Small Computer System Interface), SAS (Serial Attached SCSI), and SATA (Serial ATA).

[0062] SSD 152 includes controller 162, which integrates the electronics that bridge non-volatile memory components (e.g., NAND (NOT-AND) flash memory) to host devices (such as non-volatile memories 170a, 170b, 170n) into host 154. The controller is typically an embedded processor that executes firmware-level code and is a critical factor in SSD performance.

[0063] Controller 162 interacts with non-volatile memories 170a, 170b, and 170n via addressing function block 164. Addressing function 164 is used, for example, to manage the mapping between logical block addresses (LBAs) of corresponding physical block addresses on SSD 152 (i.e., on the non-volatile memories 170a, 170b, and 170n of SSD 152). Because the sizes of non-volatile memory pages and host sectors differ, the SSD must construct and maintain data structures that enable it to translate between the host that writes or reads data from sectors and the physical non-volatile memory pages where that data is actually placed. This table structure, or "mapping," can be constructed and maintained for sessions in the volatile memory 172 of the SSD, such as dynamic random access memory (DRAM) or some other local volatile memory component accessible by controller 162 and addressing function block 164. Alternatively, the table structure can be maintained more persistently over sessions in non-volatile memory of the SSD, such as non-volatile memory 170a, 170b-170n.

[0064] In addition to the non-volatile memories 170a, 170b-170n, addressing 164 also interacts with the data buffer cache 166. The data buffer cache 166 of the SSD 152 typically uses DRAM as a cache, similar to the cache in a hard disk drive. The data buffer cache 166 serves as a buffer or hierarchical region for transferring data to and from non-volatile memory components, and acts as a cache to speed up future requests for cached data. The data buffer cache 166 is typically implemented using volatile memory, therefore the data stored therein is not permanently stored in the cache; that is, the data is not persistent.

[0065] Finally, the SSD 152 includes one or more non-volatile memory components 170a, 170b-170n. For non-limiting purposes, the non-volatile memory components 170a, 170b-170n may be implemented as flash memory (e.g., NAND or NOR flash) or other types of solid-state memory available now or in the future. The non-volatile memory components 170a, 170b-170n are the actual memory electronics on which data is continuously stored. The non-volatile memory components 170a, 170b to 170n of the SSD 152 can be considered as an analogue to a hard disk in a hard disk drive (HDD) storage device.

[0066] Furthermore, references to data storage devices in this document may encompass multi-media storage devices (or “multi-media devices,” which may sometimes be referred to as “multi-tiered devices” or “hybrid drives”). Multi-media storage devices generally refer to storage devices that combine the functionality of a conventional HDD (see, for example, HDD 100) with an SSD (see, for example, SSD 150) that uses non-volatile memory (such as flash memory or other solid-state (e.g., integrated circuit) memory) that is electrically erasable and programmable. Because the operation, management, and control of different types of storage media typically differ, the solid-state portion of a hybrid drive may include its own corresponding controller functionality, which may be integrated with the HDD functionality into a single controller. Multi-media storage devices can be built and configured to operate and utilize the solid-state portion in a variety of ways, such as, as a non-limiting example, using the solid-state memory as cache memory, for storing frequently accessed data, for storing I / O-intensive data, for storing metadata corresponding to payload data (e.g., for assisting in decoding payload data), etc. Additionally, multi-media storage devices can be built and configured essentially as two storage devices, namely a conventional HDD and an SSD, in a single housing, with one or more interfaces for host connectivity.

[0067] Extensions and alternatives

[0068] In the foregoing description, embodiments of the invention have been described with reference to numerous specific details, which may vary depending on the specific implementation. Therefore, various modifications and changes can be made without departing from the broader spirit and scope of the embodiments. Accordingly, the invention, and the applicant's intended sole and exclusive indicator of the invention, is the set of claims in the specific form issued by this patent application, including any subsequent amendments. Any definitions of terms expressly set forth herein that are included in these claims shall determine the meaning of those terms as used in the claims. Thus, any limitations, elements, characteristics, features, advantages, or attributes not expressly cited in the claims shall not in any way limit the scope of these claims. Therefore, this specification and the accompanying drawings are to be considered illustrative rather than restrictive.

[0069] Furthermore, in this description, certain process steps may be shown in a specific order, and alphanumeric labels may be used to identify certain steps. Unless explicitly stated in the specification, the implementation is not necessarily limited to any particular order in which such steps are performed. Specifically, these labels are used only for the convenience of identifying steps and are not intended to specify or require a particular order in which such steps are performed.

Claims

1. A system comprising: A low-temperature preheater chamber is configured to absorb low-temperature waste heat from one or more relatively low-temperature electronic components and exchange the low-temperature waste heat to an organic working fluid flowing through the preheater chamber. A high-temperature evaporator, which is in fluid communication with the preheater chamber and is configured to absorb high-temperature waste heat from one or more relatively high-temperature electronic components and exchange the high-temperature waste heat to the organic working fluid flowing through the evaporator; as well as An expander, in which the heated organic working fluid is transferred from the evaporator to the expander for expansion to extract power.

2. The system of claim 1, wherein the relatively high-temperature electronic component includes one or more central processing units (CPUs) housed in an electronic device housing.

3. The system of claim 1, wherein the relatively low-temperature electronic component includes one or more data storage devices housed within an electronic device housing.

4. The system of claim 3, wherein the relatively high-temperature electronic component includes one or more central processing units (CPUs) housed together with the relatively low-temperature electronic component in the electronic device housing.

5. The system according to claim 1, wherein: The preheater chamber and the evaporator are configured to be thermally coupled to an electronic device housing, in which the relatively low-temperature electronic components and the relatively high-temperature electronic components are housed; and The system is configured such that the organic working fluid flows from the preheater chamber to the evaporator.

6. The system according to claim 5, wherein: The preheater chamber is configured to be thermally coupled to the electronic device housing via one or more first heat exchangers; and The evaporator is configured to be thermally coupled to the electronic device housing via one or more second heat exchangers.

7. The system according to claim 1, further comprising: Multiple preheater chambers, each preheater chamber being configured to be thermally coupled to a corresponding electronic device housing, wherein a corresponding relatively low-temperature electronic component is housed within the corresponding electronic device housing; Multiple evaporators, each evaporator being configured to be thermally coupled to one of the corresponding electronic device housings, a corresponding relatively high-temperature electronic component being housed in the corresponding electronic device housing, and a corresponding preheater chamber being configured to be thermally coupled to the corresponding electronic device housing; as well as A manifold configured to fluidly connect two or more evaporators in the evaporator before being fed to the expander.

8. The system according to claim 1, further comprising: A lower temperature preheater chamber, which is in fluid communication with the low temperature preheater chamber, is configured to absorb lower temperature waste heat having a lower temperature than the low temperature waste heat from one or more relatively low temperature electronic components, and exchange the lower temperature waste heat to the organic working fluid flowing through the lower temperature preheater chamber to the low temperature preheater chamber.

9. The system according to claim 1, wherein: The cryogenic electronic components are immersed in the organic working fluid within the preheater chamber; and The high-temperature electronic components are immersed in the organic working fluid in the evaporator.

10. The system according to claim 9, wherein: The cryogenic electronic components are housed within a first electronic device housing immersed in the organic working fluid within the preheater chamber; and The high-temperature electronic components are housed in a second electronic device housing that is immersed in the organic working fluid in the evaporator.

11. The system of claim 1, wherein the preheater chamber and the evaporator comprise an integrated chamber in which the cryogenic electronic components and the high-temperature electronic components are immersed in the organic working fluid.

12. The system of claim 11, wherein the cryogenic electronic component and the high-temperature electronic component are housed in a common electronic device housing immersed in the organic working fluid.

13. A method for recovering power from waste heat in a data center, the method comprising: An organic working fluid is pumped through a cryogenic preheater chamber, which is configured to exchange cryogenic waste heat from one or more relatively cryogenic electronic components with the organic working fluid flowing through the preheater chamber. The organic working fluid is pumped through a high-temperature evaporator, which is thermally coupled to the preheater chamber and configured to exchange high-temperature waste heat from one or more relatively high-temperature electronic components to the organic working fluid flowing through the evaporator. as well as The heated organic working fluid is pumped from the evaporator to the expander to generate electricity via the expansion of the organic working fluid.

14. The method of claim 13, wherein: Pumping the organic working fluid through the cryogenic preheater chamber includes passing the organic working fluid through the cryogenic electronic component, the cryogenic electronic component including one or more data storage devices housed in an electronic device housing; and Pumping the organic working fluid through the high-temperature evaporator includes passing the organic working fluid through the relatively high-temperature electronic components, which include one or more central processing units (CPUs) housed in an electronic device housing.

15. The method of claim 14, wherein: The relatively high-temperature electronic component and the relatively low-temperature electronic component are housed together in a shared electronic device housing; The preheater chamber and the evaporator are each thermally coupled to the shared electronic device housing; and Pumping the organic working fluid through the high-temperature evaporator includes pumping the organic working fluid flowing from the preheater chamber into the evaporator.

16. The method according to claim 13, further comprising: The organic working fluid is pumped from a lower temperature preheater chamber, which is thermally coupled to the cryogenic preheater chamber and configured to exchange lower temperature waste heat from one or more relatively low temperature electronic components having a lower temperature than the cryogenic waste heat to the organic working fluid flowing through the lower temperature preheater chamber.

17. The method of claim 13, wherein: Pumping the organic working fluid through the cryogenic preheater chamber includes passing the organic working fluid through the cryogenic electronic components immersed in the organic working fluid within the preheater chamber; and Pumping the organic working fluid through the high-temperature evaporator includes passing the organic working fluid through the high-temperature electronic components immersed in the organic working fluid in the evaporator.

18. The method of claim 17, wherein: The preheater chamber and the evaporator are configured as an integrated chamber, in which the low-temperature electronic components and the high-temperature electronic components are immersed in the organic working fluid; and Pumping the organic working fluid through the low-temperature preheater chamber and pumping the organic working fluid through the high-temperature evaporator includes pumping the organic working fluid through the integration chamber.

19. The method of claim 18, wherein pumping the organic working fluid through the cryogenic preheater chamber and pumping the organic working fluid through the high-temperature evaporator comprises passing the organic working fluid through a common electronic device housing that houses the cryogenic electronic components and the high-temperature electronic components and is immersed in the organic working fluid.

20. A waste heat recovery system for electronic devices, the waste heat recovery system for electronic devices comprising: A device for exchanging cryogenic waste heat from one or more relatively cryogenic electronic components to an organic working fluid, said one or more relatively cryogenic electronic components including one or more data storage devices; A device for exchanging high-temperature waste heat from one or more relatively high-temperature electronic components to the organic working fluid from the component for exchanging low-temperature waste heat, wherein the one or more relatively high-temperature electronic components include one or more central processing units (CPUs). as well as A device for expanding the heated organic working fluid from the apparatus for exchanging high-temperature waste heat in order to recover power from the low-temperature and high-temperature waste heat.

Images