Techniques for net-zero energy data processing

Abstract

A method and system for energy-optimized workload management is presented. The system includes a plurality of micro-data centers, each configured to perform computing operations that generate heat as a byproduct of the computing operations, wherein micro-data centers are geographically distributed; a thermal management module proximately disposed to each micro-data center, configured to utilize the generated heat for a non-computing application; a local monitoring module at each micro-data center configured to determine energy efficiency scores of the micro-data centers based on one or more localized energy demand conditions; and a central workload management controller communicatively coupled to the plurality of micro-data centers, the central workload management controller configured to: receive the energy efficiency scores from each micro-data center and dynamically allocate computing workloads among the plurality of micro-data centers in accordance with the received energy efficiency scores so as to maximize at least overall energy efficiency across the system.

Description

TECHNIQUES FOR NET-ZERO ENERGY DATA PROCESSINGCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of US Provisional Application No. 63 / 690,639 filed on September 4, 2024, and US Provisional Application No. 63 / 824,611 filed on June 16, 2025, the contents of which are hereby incorporated by reference.TECHNICAL FIELD

[0002] The present disclosed subject matter relates to the utilization of residual thermal energy for net-zero energy data processing.BACKGROUND

[0003] Data Center computer processing is one of the most energy intensive consumers of electricity in the world today and has been projected to be the primary consumer of electricity. It is widely known that the semiconductors used in data center processing not only consume significant amounts of electricity but also generate copious amounts of heat. Data center cooling is crucial for operating data centers and maintaining the stability and longevity of servers and other equipment. Commercially available methods include HVAC systems, which manage temperature, humidity, and air quality, and Computer Room Air Conditioning (CRAC) units, which are designed to circulate cooled air in computer rooms. In-row cooling offers localized cooling by placing units between server racks, ensuring efficient temperature control. Liquid cooling, which uses liquids to absorb and dissipate heat, is particularly effective in high- density environments. Free cooling leverages natural environmental conditions, such as outside air, to reduce reliance on mechanical systems, thereby saving energy.

[0004] Chilled water cooling is another widely used method, involving the circulation of chilled water through heat exchangers to cool air, making it effective for large-scale data centers. Each method is chosen based on the specific requirements of the data center, including its size, location, and energy efficiency goals, to ensure optimal operation and minimize environmental impact.

[0005] Given the enormous amount of energy required for cooling and the growth in demand for data processing for such data centers, which have both negative energy consumption and environmental consequences, there is a critical need to address the significant energy consumption and associated environmental impact of these facilities. Data centers are vital for supporting the world's digital infrastructure, but they also contribute significantly to global energy use. A considerable portion of this energy, up to 40% in some cases, is consumed by cooling systems designed to prevent overheating of servers and other critical equipment.

[0006] As data centers continue to grow in both number and size, their energy demands have risen, making them one of the key contributors to global carbon emissions. This is particularly concerning as the world moves toward achieving net- zero carbon targets by 2050. Traditional cooling methods, such as mechanical chillers, are not only energy-intensive but also reliant on fossil fuels, which exacerbates the carbon footprint of data centers.

[0007] Additionally, there are challenges with developing infrastructure and aligning heat supply with demand to overcome hurdles associated with recovering and reusing residual heat generated by data centers.

[0008] Therefore, it would be advantageous to provide a solution that would cure the deficiencies noted above.SUMMARY

[0009] A summary of several example embodiments of the disclosure follows. This summary is provided for the convenience of the reader to provide a basic understanding of such embodiments and does not wholly define the breadth of the disclosure. This summary is not an extensive overview of all contemplated embodiments and is intended to neither identify key or critical elements of all embodiments nor to delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more embodiments in a simplified form as a prelude to the more detailed description that is presented later. For convenience, the term “some embodiments” or “certain embodiments” may be used herein to refer to a single embodiment or multiple embodiments of the disclosure.

[0010] A system of one or more computers can be configured to perform particular operations or actions by virtue of having software, firmware, hardware, or a combination of them installed on the system that in operation causes or cause the system to perform the actions. One or more computer programs can be configured to perform particular operations or actions by virtue of including instructions that, when executed by data processing apparatus, cause the apparatus to perform the actions.

[0011] In one general aspect, a system may include a plurality of micro-data centers, each configured to perform computing operations that generate heat as a byproduct of the computing operations, where the plurality of micro-data centers are geographically distributed. The system may also include a thermal management module proximately disposed to each micro-data center, configured to utilize the generated heat for a noncomputing application. The system may furthermore include a local monitoring module at each micro-data center of the plurality of micro-data centers configured to determine energy efficiency scores of the plurality of micro-data centers based on one or more localized energy demand conditions. The system may in addition include a central workload management controller communicatively coupled to the plurality of micro-data centers, the central workload management controller configured to: receive the energy efficiency scores from each of the plurality of micro-data centers and dynamically allocate computing workloads among the plurality of micro-data centers in accordance with the received energy efficiency scores so as to maximize at least overall energy efficiency across the system. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0012] In one general aspect, the method may include performing computing operations at a plurality of micro-data centers, each micro-data center generating heat as a byproduct of the computing operations, where the plurality of micro-data centers are geographically distributed. The method may also include determining, at each micro-data center of the plurality of micro-data centers, an energy efficiency score of the micro-data center based on one or more localized energy demand conditions. The method may furthermore include transmitting the energy efficiency scores from the plurality of micro-data centers to a central workload management controller. Themethod may in addition include dynamically allocating computing workloads among the plurality of micro-data centers, by the central workload management controller, for a non-computing application in accordance with the determined energy efficiency scores so as to maximize at least overall energy efficiency of the plurality of micro-data centers. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0013] In one general aspect, a micro-data center may include a computing module configured to perform data computing operations and to generate heat as a byproduct of the data computing operations. A micro-data center may also include a thermal management module integrated with the computing module and configured to utilize at least a portion of the generated heat to provide a heating or cooling function to an environment proximate to the micro-data center and to the micro-date center. Center may furthermore include a communications interface configured to enable the microdata center to exchange data with external computing networks, where the micro-data center is configured as a self-contained system such that both computing services and heating or cooling services are provided from a single integrated product. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0014] In one general aspect, a system may include a computing device configured to generate heat during operation. The system may also include a heat transfer medium configured to absorb thermal energy from the computing device. The system may furthermore include a transport conduit thermally coupled to the computing device and configured to convey the thermal energy carried by the heat transfer medium from a first location adjacent to the computing device to a second location remote from the computing device. The system may in addition include a secondary energy utilization system located at the second location and configured to receive the thermal energy carried by the heat transfer medium and utilize the thermal energy to perform a non- computational process. Other embodiments of this aspect include correspondingcomputer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The foregoing and other objects, features, and advantages of the disclosure will be apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0016] Figure 1 shows an example block diagram of an energy-optimized workload management system 100 dispersed across a plurality of premises in accordance with some exemplary embodiments of the present disclosure.

[0017] Figure 2 shows a block diagram of a premise that utilizes energy-optimized workload management according to an embodiment.

[0018] Figure 3 illustrates an example flowchart of a process for energy-optimized workload management according to an embodiment.DETAILED DESCRIPTION

[0019] The embodiments disclosed herein are only examples of the many possible advantageous uses and implementations of the innovative teachings presented herein. In general, statements made in the specification of the present application do not necessarily limit any of the various claimed embodiments. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise indicated, singular elements may be plural and vice versa with no loss of generality. In the drawings, like numerals refer to like parts through several views.

[0020] The technical problem addressed by the disclosed subject matter is the high energy consumption of data centers, primarily due to the need for computing and cooling, which is driven by the significant heat emitted by computers. The objective of the disclosed embodiments is to decentralize data centers into a plurality of smaller, distributed locations (micro-data centers), allowing the residual heat generated by computing equipment to be repurposed for cooling and / or heating the micro-data centers and an environment proximate to these micro-data centers by utilizing liquid desiccant air conditioning systems and / or by existing commercially available HVAC installation. The micro-data centers are deployed in different premises.

[0021] One technical effect of the disclosed subject matter is the significant energy savings and reduction in carbon emissions. In some exemplary embodiments, the present disclosure features a local monitoring module configured to interface with a modified commercially available HVAC (thermal management module) of a premise to utilize the residual heat generated by the computing elements of the micro-data centers.

[0022] Micro-data centers are defined as geographically distributed computerized equipment such as, but not limited to, computer servers, storage systems, networking equipment, one or more Graphics Processing Units (GPU), one or more storage devices, any combination thereof, and the similar equipment configured to process large data volumes and support complex, high-intensity operations in applications including, but not limited to, cloud computing; big data analytics; Al processing including training and inference of machine learning models, language models, and the like; content delivery; enterprise operations; a combination thereof; and the like. In some exemplary embodiments, the micro-data centers are geographically dispersed across different locations and may be linked together (interconnected) via various communication channels including, but not limited to, fiberoptic trunks (e.g., using a cloud router), wireless connections (e.g., through satellite or other wireless transmission systems), a combination thereof, and the like. In some embodiments, It should be noted that micro-data centers may be governed by a central workload management controller (hereinafter, central workload management controller or central workload management router) using the various communication channels.

[0023] In one embodiment, the system includes a mechanism for capturing thermal energy generated by one or more computing components during operation, and for repurposing that energy to drive processes external to the computing function. The mechanism employs a heat transfer medium, such as a liquid desiccant or an alternative fluid medium with suitable thermodynamic properties, that is capable of absorbing, retaining, and transporting heat from the location of generation (e.g., a central processing unit, graphics processor, or memory module) to a remote location. At this secondary location, the thermal energy is released and utilized to power a non-computational process, such as environmental cooling, dehumidification, chemical processing, or other forms of energy recovery and reuse.

[0024] In a further embodiment, the captured thermal energy, collected using the mechanism described above, is employed as a source of energy to activate or facilitate a cooling process targeted at the computing components themselves. For example, the thermal energy may drive a thermally powered absorption chiller, evaporative cooler, or other active cooling system that removes heat from the original computing device. This creates a closed-loop or semi-closed-loop thermal management cycle in which the system reuses a portion of the heat the system generates to enhance the system’s own cooling performance, thereby improving energy efficiency and reducing reliance on external power sources for cooling.

[0025] FIG. 1 shows an example block diagram of an energy-optimized workload management system 100 dispersed across a plurality of premises 110, in accordance with some exemplary embodiments of the present disclosure.

[0026] It should be noted that a commercially available data center is a facility that incorporates computerized equipment, such as, but not limited to, computer servers, storage systems, and networking equipment, used for storing, managing, processing, and distributing large volumes of data. The computerized equipment is essential for processing large data volumes and supporting complex, high-intensity operations in applications including, but not limited to, cloud computing; big data analytics; Al processing, including training and inference of machine learning models, language models, and the like; content delivery; enterprise operations; a combination thereof; and the like.

[0027] In an embodiment, Al processing provides the foundation for artificial intelligence (Al) systems, including large language models (LLMs), in both the training and inference phases. In training, vast and heterogeneous datasets are collected, processed, and transformed using distributed storage and computation frameworks, enabling machine learning models, such as deep neural networks and LLMs, to be trained across massive corpora of text, code, and multimodal data with parallelized optimization of billions of parameters. In inference, big data analytics supports the deployment of trained models by supplying high-throughput, low-latency pipelinescapable of handling real-time or near real-time data streams, allowing models to generate predictions, classifications, natural language responses, or recommendations at scale. For LLMs, this includes tokenization, context management, and efficient distributed inference across specialized hardware. Together, these capabilities enable Al systems to leverage big data infrastructure for the computationally intensive optimization of models during training and the efficient, scalable execution of models, including LLM-based processing, during inference.

[0028] Commercially available data centers include, but are not limited to, processing servers, storage systems (e.g., hard drives, SSDs), networking equipment (switches, routers, firewalls), and redundant power supplies (generators, UPS). Commercial available data centers often include cooling equipment, cooling apparatuses, a combination thereof, and the like to cool the processing servers, storage systems, and the like (listed above). Typically, data centers are owned by a single organization or operated by third-party providers offering colocation or cloud services.

[0029] In some exemplary embodiments, energy-optimized workload management system 100 may include a Backbone-Cloud 120 and a plurality of micro-data centers. In some exemplary embodiments, the Backbone-Cloud 120 and the micro-data centers of each premise 110 may be interconnected by communication channels 121. Additionally, or alternatively, a central workload management controller 130 may be also interconnected by communication channels 121 to Backbone-Cloud 120 and to the micro-data centers of premises 110.

[0030] In some embodiments, the plurality of micro-data centers are configured as modular, distributed units operable within residential or commercial environments. In some embodiments, the energy-optimized workload management system 100 maximizes overall energy reuse by routing computing workloads to micro-data centers located in premises 110 in geographic regions having active heating or cooling demands.

[0031] In some embodiments, Backbone-Cloud 120 may utilize a network architecture featuring a distributed system of routers and networking equipment. This setup may be managed by a task management server (not shown) that distributes tasks, such asstoring, managing, processing, and routing large volumes of data, across the various micro-data centers of the energy-optimized workload management system 100.

[0032] In some exemplary embodiments, Backbone-Cloud 120 may include a distributed infrastructure of routers and switches across various locations to manage data traffic. Additionally, or alternatively, Backbone-Cloud 120 dynamically adjusts to handle varying data volumes and traffic patterns while managing load balancing among the micro-data centers, considering the heat load and dissipation of each micro-data center.

[0033] It should be noted that the Backbone-Cloud 120 may be utilized to connect the micro-data centers to external networks and the internet, enabling remote management, data access, and communication between energy-optimized workload management system 100 and external resources. This connectivity enhances the energy-optimized workload management system 100's functionality by facilitating realtime data exchange and control over various components.

[0034] In some exemplary embodiments, Backbone-Cloud 120 may enable centralized management for streamlined configuration, monitoring, and maintenance by central workload management controller 130.

[0035] Premise 110 may refer to a real estate location having a HVAC installation including, but not limited to, residential, commercial, industrial, or institutional buildings (such as offices, shopping malls, warehouses, and production facilities); a heatemitting transport system (such as a car, truck, bus, and train); and any combination thereof, and similar structures. In some exemplary embodiments, premise 110 may accommodate a segment of computerized equipment that constitutes a portion of the overall computerized equipment of a commercially available data center.

[0036] Communication channel 121 allows for fast and reliable communication between the Backbone-Cloud 120 and the micro-data centers of the plurality of premises 110. In some exemplary embodiments, communication channels 121 have the ability to handle large data loads, with minimal signal loss over long distances, low latency, high reliability, high durability, and high bandwidth.

[0037] It should be noted that the Backbone-Cloud 120 and communication channel 121 is one example for inter-communication between micro-data centers. Othertransport media may include, but is not limited to, carrier-provided Ethernet private lines (EPL / EVPL), which extend Layer 2 connectivity across facilities with guaranteed bandwidth; Multi-Protocol Label Switching (MPLS) or IP-VPN services, which enable resilient Layer 3 connectivity with traffic engineering across wide-area networks; satellite-based communication links, including low-earth orbit constellations, which provide backup or connectivity in geographically remote locations; and cloud provider interconnects, such as AWS® Direct Connect, Microsoft® Azure ExpressRoute, or Google® Cloud Interconnect, which offer high-bandwidth, private connections between enterprise or colocation data centers and cloud data centers.

[0038] Central workload management system 130 may function as a computing system that oversees and controls dispersed computing resources from a centralized platform. Central workload management controller 130 offers unified control, optimizes resource allocation, enables real-time monitoring and maintenance, ensures consistent security and compliance, and supports scalability. By providing a streamlined and coordinated management framework, central workload management controller 130 enhances the efficiency, security, and scalability of the energy-optimized workload management system 100 described in the present disclosure.

[0039] The central workload management controller 130 is communicatively coupled to the micro-data centers of the premises 110. In some embodiments, the central workload management controller 130 is configured to receive energy efficiency scores from the micro-data centers and dynamically allocate computing workloads among the plurality of micro-data centers in accordance with the received energy efficiency scores so as to maximize overall energy reuse across the energy-optimized workload management system 100.

[0040] In some embodiments, the energy efficiency score is determined on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse. In some embodiments, the energy efficiency score includes a factor based on the source of electrical energy powering the respective micro-data center. For example, a renewable or green energy source contributes a higher score relative to a non-renewable energy source.

[0041] Additionally, in some embodiments, the central workload management controller 130 is configured to consider temporal variations in local energy demand, including seasonal heating and cooling requirements, as well as meteorological events, when allocating workloads between the plurality of micro-data centers. For example, the central workload management controller 130 is configured to allocate larger computing workloads to micro-data centers of premises located in geographically cooler conditions than to micro-data centers of premises located in geographically warmer conditions. Meteorological events can be received through an API or other means from meteorological information services.

[0042] In some embodiments, the central workload management controller 130 is configured to configured to determine when a building proximate to at least one microdata center of the plurality of micro-data centers requires heating. According to this embodiment, when it is determined that a building proximate to the at least one microdata center requires heating, the central workload management controller 130 is configured to operate the at least one micro-data center. When it is determined that a building proximate to the at least one micro-data center does not require heating, the central workload management controller 130 is configured to stop any operations of the at least one micro-data center. In further embodiments, the central workload management controller 130 is configured to allocate workloads based on the energy efficiency scores of operating micro-data centers.

[0043] Figure 2 shows a block diagram of a premise that utilizes energy-optimized workload management according to an embodiment. In some exemplary embodiments, premise 110, which incorporates a commercially available HVAC installation, may accommodate at least one micro-data center 210 and a local monitoring module 220, which is configured to interface between Backbone-Cloud 120, micro-data center 210, and a modified HVAC system (hereinafter, HVAC System 230 or thermal management module 230), and media devices 260.

[0044] In some exemplary embodiments, micro-data center 210 may include at least one computerized equipment including, but not limited to, computer servers, storage systems, networking equipment typically used in commercially available data centers, one or more Graphics Processing Units (GPU), one or more storage devices, and anycombination thereof, and the like. The computerized equipment of micro-data center 210 of one premise 110 may be used in conjunction with micro-data center 210 of other premises 110 (e.g., of energy-optimized workload management system 100, FIG. 1) for storing, managing, processing, and distributing large volumes of data.

[0045] Micro-data center 210 may be utilized to process large data volumes and support complex, high-intensity operations in applications including, but not limited to, cloud computing; big data analytics; cryptocurrency mining; Al processing including training and inference of machine learning models, language models, and the like; content delivery; enterprise operations; a combination thereof; and the like. It should be noted that the Backbone-Cloud 120 and fiberoptic trunk 121 is one example for inter micro-data center communication. Othe transport media may include carrier-provided Ethernet private lines (EPL / EVPL), which extend Layer 2 connectivity across facilities with guaranteed bandwidth; Multi-Protocol Label Switching (MPLS) or IP-VPN services, which enable resilient Layer 3 connectivity with traffic engineering across wide-area networks; satellite-based communication links, including low-earth orbit constellations, which provide backup or connectivity in geographically remote locations; and cloud provider interconnects, such as AWS Direct Connect, Microsoft Azure ExpressRoute, or Google Cloud Interconnect, which offer high-bandwidth, private connections between enterprise or colocation data centers and cloud data centers.

[0046] In some embodiments, the micro-data center 210 incorporates software that enables the micro-data center 210 to function as part of a server farm. Micro-data center 210 (of energy-optimized workload management system 100, FIG. 1 ) that is situated in premise 110 may utilize a Backbone-Cloud 120 for connectivity with any server farm.

[0047] In some exemplary embodiments, the at least one piece of computerized equipment may include processing servers, storage systems (e.g., hard drives, SSDs), networking equipment (e.g., switches, routers, firewalls), and Uninterruptible Power Supplies (UPS) that may be supported by batteries and / or generators.

[0048] The thermal management module 230 may be based on the existing HVAC installation and equipment of premise 110, which is upgraded to manage the cooling ofthe micro-data center 210 and / or to utilize the heat emitted by the micro-data center 210.

[0049] In some exemplary embodiments, upgrading a commercially available HVAC of Premise 110 to the thermal management module 230 may involve adding at least one air conditioning unit to increase the cooling capacity of the existing HVAC installation to meet the cooling needs of the micro-data center 210 and the environment proximate (e.g., premises 110) to the micro-data center 210 during relatively warmer periods. Additionally, or alternatively, upgrading may include additional ventilation units that allow for the introduction of cold air from outside premise 110 during relatively cooler periods. In some exemplary embodiments, upgrading may include routing heat emitted from the micro-data center 210 to heat premise 110 during relatively cooler periods, thereby leveraging residual heat for heating the premise 110 during such periods. The routing of heat emitted by the micro-data center 210 may be realized by a heat pump adapted to transfer heat from micro-data center 210 to the other areas in the premise 110 that require heating, thereby cooling the micro-data center 210.

[0050] In some exemplary embodiments, thermal management module 230 includes a heat-to-cool system (not shown) that utilizes heat generated by micro-data center 210 during operation for cooling the micro-data center 210. The heat-to-cool system may include absorption chillers, thermoelectric coolers, and any combination thereof, or similar technologies that can convert waste heat into cooling.

[0051] In such an embodiment, instead of simply dissipating the heat generated by micro-data center 210 into the environment, thermal management module 230 channels the waste heat into the heat-to-cool system, which uses the heat energy to drive a cooling process. Essentially, the waste heat is converted into a cooling effect.

[0052] It should be noted that in this embodiment, the thermal management module 230 not only cools the processors with the conventional AC unit but also enhances overall cooling by using the waste heat through the heat-to-cool system.

[0053] It should also be noted that the heat-to-cool system in thermal management module 230 increases the overall efficiency by repurposing waste heat to generate cooling, rather than simply removing it, thereby reducing the amount of energy needed to power the cooling process.

[0054] Moreover, the waste heat generated by micro-data center 210 produces additional cooling that can further cool the surrounding space in premise 110, making the entire system more efficient.

[0055] In some exemplary embodiments, upgrading a commercially available HVAC of premise 110 to thermal management module 230 may include an additional thermal sensor (not shown) positioned in close proximity to the micro-data center 210 for measuring temperature of the micro-data center 210 and premises 110. Additionally, or alternatively, upgrading may incorporate an interface configured to facilitate communication between the thermal management module 230 and the local monitoring module 220.

[0056] It should be appreciated that at least one of the subsystems in diagram 200 (hereinafter, subsystems) may be integrated within the thermal management module 230. In some embodiments, these subsystems are mechanically and thermally connected to thermal management module 230.

[0057] In such embodiments, the thermal management module 230 physically incorporates one or more subsystems that allow for seamless interaction between the heating / cooling system and other home automation and data processing components.

[0058] Thermal management module 230 may utilize heat pump and / or liquid desiccant technologies to provide heating and cooling, optimizing energy efficiency in various premises year-round. In some embodiments, heat pump technology transfers heat to warm or cool interiors, functioning without generating new heat, and adapts to different settings such as residential and commercial environments. It includes types like air source, ground source, water source, and ductless mini-splits, and can be combined with other technologies like gas boilers to enhance energy efficiency. Additionally, or alternatively, liquid desiccant technology controls humidity and cools air by absorbing moisture with a chemical solution such as lithium chloride, calcium chloride, or the like. This process is thermally activated, using waste heat or solar heat, and the desiccant is regenerated with thermal energy from the heat pump, boosting system performance and energy conservation.

[0059] In some embodiments, the subsystems may be integrated and thermally connected to the one or more thermal management modules 230. Thermal connectivitybetween subsystems and thermal management module 230 leverages heat from electronic devices in an integrated system. Specifically, the thermal management module 230 utilizes the heat generated by subsystems, such as GPUs and other electronic devices as a resource for heating, thereby cooling these devices and also reducing the energy load on the heating system. This dual functionality is crucial in reducing overall energy consumption and improving the efficiency of the micro-data center.

[0060] In some embodiments, thermal connectivity may be realized by placing heat exchangers near high-heat devices like GPUs, which capture excess heat and transfer it via a fluid medium to thermal management module 230 for heating during relatively colder periods.

[0061] In some embodiments, thermal connectivity may be achieved through thermally conductive piping that transports heat from electronic devices to thermal management module 230, redirecting heat from areas like server rooms to water heaters or radiators.

[0062] In some embodiments, thermal connectivity may be realized by custom ductwork designed to funnel warm air from electronic devices to the air handling units of thermal management module 230, effectively redistributing heat in large spaces.

[0063] In some embodiments, thermal connectivity may be realized by integrating cooling and heating loops with electronic devices, which are connected to the building’s heating loop, allowing heat absorbed by cooling agents to be used by a heat pump or boiler to warm the building.

[0064] In some embodiments, heat generated by processors or GPUs during data processing is directly routed to assist in driving the HVAC functions, particularly heating during colder periods. This process reclaims what would otherwise be waste thermal energy and channels the thermal energy via mechanisms including, but not limited to, integrated ducts, physical channels, heat exchangers, a combination thereof, and the like to offset the energy demand of the HVAC unit. As a result, the system benefits from dual-purpose computing — performing data operations while also passively supporting environmental climate control within the premises.

[0065] In some embodiments, thermal connectivity may be realized by ventilation systems that channel warm air from the premises to thermal collectors in thermal management module 230, enhancing heat recovery efficiency.

[0066] The local monitoring module 220 may be a computerized apparatus designed to manage communication between the Backbone-Cloud 120 and the micro-data center 210. Additionally, or alternatively, local monitoring module 220 may control the thermal management module 230to regulate the temperature of the micro-data center 210 and premise 110, ensuring a stable climate environment for proper operation of micro-data center 210 and utilizing residual heat for other purposes within premises 110. In some embodiments, local monitoring module 220 and the sub-components of local monitoring module 220 are integrated into micro-data center 210. In some embodiments, thermal management module 230 is integrated into micro-data center 230. In some embodiments, thermal management module 230 includes, but is not limited to, a cooling module and a thermally-driven cooling system (such as an absorption chiller).

[0067] In some embodiments, the local monitoring module 220 is managed by an energy management software (EMS) configured to operate the thermal management module 230 while coordinating it with other energy-consuming activities. The EMS causes the local monitoring module 220 to dynamically adjust energy use, enhancing the efficiency of the thermal management module 230 by reducing or increasing its activity based on the overall energy requirements and availability, thus optimizing the premise's energy consumption.

[0068] In some exemplary embodiments, local monitoring module 220 may include a computing module. Computing module includes, but is not limited to, at least a Central Processing Unit (CPU) 221 , a Memory 222, a Network Interface Card (NIC) 224 (hereinafter, communications interface 224 or Network Interface Card 224), and an Input / Output (I / O) Module 223, and a storage device (not shown).

[0069] CPU 221 may be a Central Processing Unit (CPU), a microprocessor, an electronic circuit, an Integrated Circuit (IC), or the like. Additionally, or alternatively, local monitoring module 220 can be implemented as firmware written for or ported to a specific processor such as Digital Signal Processor (DSP), Graphical Processing Unit(GPU), Tensor processor, or microcontrollers or can be implemented as hardware or configurable hardware such as field programmable gate array (FPGA) or application specific integrated circuit (ASIC). CPU 221 may be utilized to perform computations required by local monitoring module 220 or any of its subcomponents.

[0070] In some exemplary embodiments, local monitoring module 220 may include a Memory 222. Memory 222 may be comprised of volatile and / or non-volatile memories, based on technologies such as semiconductor, magnetic, optical, flash, a combination thereof, or the like. For example, Memory 222 can be a Flash disk, a Random Access Memory (RAM), a memory chip, an optical storage device such as a CD, a DVD, or a laser disk; a magnetic storage device such as a tape, a hard disk, storage area network (SAN), a network attached storage (NAS), or others; a semiconductor storage device such as Flash device, memory stick, or the like.

[0071] In some exemplary embodiments, Memory 222 may store program code that activates CPU 221 to perform tasks associated with communication between the Backbone-Cloud 120 and the micro-data center 210, as well as controlling the thermal management module 230. The program code may be implemented as one or more sets of interrelated computer instructions, executed by CPU 221 or another processor. These components may be organized as executable files, dynamic libraries, static libraries, methods, functions, services, or similar constructs, and can be programmed in any programming language and operate under any computing environment.

[0072] In some exemplary embodiments, I / O Module 223 may be utilized by local monitoring module 220 as an interface to communicate information and instructions with the thermal management module 230 and to acquire temperature measurement information from the thermal sensor (not shown). Additionally, or alternatively, I / O Module 223 may provide an interface for users of premises 110 (not shown), such as by delivering outputs, visualized results, reports, or similar information.

[0073] In some embodiments, acquiring temperature measurement information from the thermal sensor is used to compute energy efficiency scores of micro-data centers 210 of premises 110. As explained above, energy efficiency scores are based on one or more localized energy demand conditions. Energy efficiency scores represent a measure of potential reuse of the generated heat of micro-data center 210 for a non-computing application (e.g., heating and / or cooling premise 110). In some embodiments, the energy efficiency score is determined on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse. In some embodiments, the energy efficiency score further includes a factor based on the source of electrical energy powering the respective micro-data center 210. For example, a renewable or green energy source contributes a higher score relative to a non-renewable energy source.

[0074] In some embodiments, the central workload management controller 130, FIG. 1. is configured to receive energy efficiency scores from the micro-data centers 210 and dynamically allocate computing workloads among the plurality of micro-data centers 210 in accordance with the received energy efficiency scores so as to maximize overall energy reuse across the energy-optimized workload management system 100, FIG. 1. Additionally, as explained above, in some embodiments, the central workload management controller 130, FIG. 1 is configured to consider temporal variations in local energy demand, including seasonal heating and cooling requirements, as well as meteorological events, when allocating workloads between the plurality of micro-data centers 210. For example, the central workload management controller 130, FIG. 1 is configured to allocate larger computing workloads to micro-data centers 210 of premises 110 located in geographically cooler conditions than to micro-data centers 210 of premises 110 located in geographically warmer conditions.

[0075] In some embodiments, the controller computes an energy efficiency score on a normalized zero-to-ten (0-10) scale by (i) determining a reuse component based on an energy reuse ratio and / or energy reuse effectiveness derived from metered energy delivered to useful secondary loads, and (ii) determining a source component based on at least one of renewable supply fraction and carbon intensity of the electrical energy. The central workload management controller 130, FIG. 1 combines the components via a weighted function, optionally with diminishing-returns shaping, availability adjustments, and temporal aggregation, to yield a score of 0 for no recoverable reuse with carbon-intensive supply and 10 for complete recoverable reuse with fully renewable or near-zero-carbon supply.

[0076] In some embodiments, the central workload management controller is configured to determine when a building proximate to at least one micro-data center of the plurality of micro-data centers requires heating. According to this embodiment, when it is determined that a building proximate to the at least one micro-data center requires heating, the central workload management controller is configured to operate the at least one micro-data center. When it is determined that a building proximate to the at least one micro-data center does not require heating, the central workload management controller is configured to stop any operations of the at least one microdata center. In further embodiments, the central workload management controller is configured to allocate workloads based on the energy efficiency scores of operating micro-data centers.

[0077] In some exemplary embodiments, the instructions communicated by local monitoring module 220 may include hardware signals used for regulating temperature and activating ventilation units to allow outdoor air to assist in cooling the micro-data center 210. And activating additional HVAC units or utilizing residual heat to warm other parts of the premises.

[0078] In some embodiments, the I / O module 223, utilized by the local monitoring module 220, interfaces with smart thermal sensors installed on one or more thermal management modules 230 and subsystems. These sensors monitor temperature fluctuations near heat-generating devices. The real-time data acquired from these sensors is managed by the EMS. Utilizing this data, the local monitoring module 220 orchestrates the control of thermal management module 230, allowing for dynamic adjustments in its operations based on the real-time heat output. This functionality enables the thermal management module 230 to either ramp up or scale down its heating capacity, thereby optimizing energy consumption in response to surplus heat from electronic devices.

[0079] In some embodiments, the system 200 integrates additional sensors and control mechanisms that fine-tune the operation of the thermal management module 230 based on comprehensive real-time data from all integrated devices. This sophisticated approach not only ensures optimal climate control within the premises but also harmonizes energy expenditure. During periods when electronic devices generateexcess heat, the system strategically reduces the operational demands on the thermal management module 230, leveraging this waste heat effectively and reducing overall energy costs.

[0080] In some embodiments, sensors embedded within the system 200 may utilize sensors configured to detect ambient human presence via heat signatures, room-wide temperature gradients, sound for voice control commands, and real-time humidity. These multimodal inputs feed into the local monitoring module 220 and the EMS software, allowing for context-optimized HVAC adjustments such as reducing airflow in unoccupied rooms, increasing dehumidification during high-moisture events, or activating cooling in response to voice prompts. This adaptive behavior further improves operational efficiency and user comfort.

[0081] It should be emphasized that by efficiently managing heat production and usage, the thermal management module 230 contributes to a significant reduction in energy consumption, which in turn lowers the carbon footprint associated with heating and cooling the premises. The integrated approach of the present disclosure enhances energy efficiency and leverages the thermal outputs of electronic devices, i.e., subsystems, aligning with energy-efficient and cost-effective solutions. This confirms the relevance of the thermal connection strategies of this invention.

[0082] The Network Interface Card (NIC) 224 may be a hardware component configured to enable and manage Ethernet communication between micro-data center 210 and the Backbone-Cloud 120 via Fiberoptic-Trunk 121. In some exemplary embodiments, NIC 224 may be used to exchange information and instructions related to tasks assigned by the central workload management system 130, FIG. 1. Additionally, or alternatively, NIC 224 may report monitored temperatures to the central workload management controller 130, FIG. 1. for use in task scheduling and load balancing.

[0083] In some embodiments, NIC 224 may be utilized to form a Local Area Network (LAN) (or a premise local network) by connecting devices such as micro-data center 210, local monitoring module 220, and at least one media device 260. Additionally, NIC 224 connects the premise local network to Backbone-Cloud 120, i.e., external networks, typically the internet, and serves as a central access and management pointfor data flow within the premise local network. NIC 224 also enhances security, manages network traffic, and connects various electronic devices.

[0084] In some embodiments, NIC 224 incorporates a wireless access point, allowing wireless devices to connect to the premise local network and access the internet. Additionally, or alternatively, NIC 224 facilitates the connection of various home devices, including computers, smart TVs, mobile devices, and smart home devices such as thermostats. In some embodiments, NIC 224 supports media sharing across the premise local network, enabling users to stream videos, music, and other media content from one device to another.

[0085] In some embodiments, the system includes a user-facing display configured to serve both interactive and passive functions. The display may present system controls, environmental metrics, and user-tailored media. Additionally, or alternatively, during idle states or ambient operation, the display may serve as an advertising screen to display targeted content. In some embodiments, revenue from such advertising may subsidize the overall cost of the integrated system, contributing to affordability and user engagement.

[0086] In some embodiments, NIC 224 may be utilized to form an Internet of Things (loT) gateway that acts as a central hub, incorporating multiple loT-based modules that facilitate, automate, and optimize the operation of appliances such as window shutters, laundry machines, and dishwashers within the premises. The loT gateway is configured to aggregate data, translate protocols, and secure transmissions from loT- based modules within loT networks. Additionally, or alternatively, loT gateway processes data locally through edge computing, enables remote management of loT units, and supports connectivity options like Wi-Fi and Bluetooth to enhance network integration to devices including, but not limited to, an entertainment system, HVAC system, smart appliance, industrial equipment, telecommunications device (e.g., switch, route, edge device, etc.), consumer electronic device, media devices 260, a combination thereof, and the like.

[0087] In some embodiments, media devices 260 encompass a variety of electronic entertainment equipment, including smart TVs, audio players, game consoles, Blu-ray / DVD players, stereo systems, streaming devices, home theater systems, and media storage devices, any combination thereof, and the like.

[0088] It should be noted that media devices 260 can utilize storage options such as hard drives, network-attached storage (NAS), and cloud services that also feature enhanced connectivity through Wi-Fi, Bluetooth, and USB. In some embodiments, such media devices 260 are integrated in micro-data center 210 and the energy-optimized workload management system is configured to utilize residual heat generated from the media devices 260.

[0089] In additional embodiments, residual heat generated from heat-emitting transport systems (not shown) including, but not limited to, a car, a truck, a train, a combination thereof, and the like is utilized to perform cooling and / or heating functions to micro-data center 210. As such, the heat source for a micro-data center 210 may be from a stationary or mobile source.

[0090] It will be appreciated that the present disclosure involves an innovative approach to cooling micro-data center 210 by integrating them with thermal management module 230 or leveraging natural cold environments, depending on the geographic location and climate conditions.

[0091] According to one aspect of the present disclosed subject matter, the thermal management module 230 outlet installed alongside micro-data center 210 enables the cooling of both the micro-data center 210 and the premise 110 space. In summer, as the thermal management module cools the premises 110, it simultaneously cools the micro-data center 210, creating a synergy that optimizes energy use.

[0092] In cold climate utilization, instead of using energy-intensive methods to cool the micro-data center 210, the system uses naturally cold ambient air, simultaneously heating the house and cooling the micro-data center 210, further enhancing energy efficiency.

[0093] As explained in more detail above, according to another aspect of the present disclosed subject matter, the central workload management system 130 oversees multiple micro-data centers 210 dispersed across various geographic locations, shifting the computational load based on factors like time of day, season, and energy costs.For instance, during January, more processing could be allocated to servers in the northern hemisphere, and in August, to the southern hemisphere.

[0094] According to yet another aspect of the present disclosed subject matter, central workload management controller 130, FIG. 1 also considers outdoor temperature and humidity, prioritizing the operation of servers in locations where the environmental conditions are most favorable, further optimizing energy usage, and minimizing the carbon footprint.

[0095] It should also be appreciated that the central workload management controller 130, FIG. 1 , which dynamically distributes the computational load and utilizes environmental conditions, can significantly reduce carbon emissions and lower energy consumption in data center operations. By optimizing load distribution and employing natural cooling methods, it not only enhances energy efficiency and reduces greenhouse gas emissions but also offers substantial cost savings. This concept has the potential to revolutionize data center operations, emphasizing energy efficiency and environmental sustainability.

[0096] According to yet another aspect of the present disclosed subject matter, an HVAC manufacturer could produce and sell a “generic” local monitoring module 220 as part of a commercially available HVAC system, including additional retrofits that upgrade the system to operate like the thermal management module 230 described in the present disclosure. In such applications, the owner of premises 110 could "rent" or "sell" its services to a company that provides data center services.

[0097] FIG. 3 illustrates an example flowchart of a process 300 for energy-optimized workload management according to an embodiment.

[0098] At S310, computing operations are performed at micro-data centers.

[0099] The computing operations performed at a plurality of micro-data centers generate heat as a byproduct of the computing operations. Each micro-data center may generate more or less heat as a byproduct of the computing operations than other micro-data centers.

[0100] In some embodiments, other computing devices of a premise that are independent from, coupled to, or integrated in the micro-data center including, but not limited to, an entertainment system, HVAC system, smart appliance, industrialequipment, telecommunications device, consumer electronic device, a combination thereof, and the like are configured to generate heat as a result of performing computing operations.

[0101] Micro-data centers perform compute operations for various purposes including, but not limited to, cloud computing; big data analytics; Al processing including training and inference of machine learning models, language models, and the like; content delivery; enterprise operations; a combination thereof; and the like.

[0102] At S320, an energy efficiency score of the micro-data centers is determined. In some embodiments, determining an energy efficiency score is based on one or more localized energy demand conditions. The energy efficiency score represents a measure of potential reuse of the generated heat for a non-computing application including, but not limited to, providing heating to a building proximate to at least one of the plurality of micro-data centers, providing cooling via a thermally-driven cooling system, a combination thereof, and the like.

[0103] In some embodiments, determining the energy efficiency score includes assigning a value on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse. In some embodiments, determining the energy efficiency score includes incorporating a factor based on the source of electrical energy powering the respective micro-data center. For example, a renewable or green energy source contributes a higher energy efficiency score relative to a non-renewable energy source.

[0104] At S330, the energy efficiency score is transmitted to a central workload management controller. The central workload management controller is communicatively coupled to the micro-data centers of the premises.

[0105] In some embodiments, the central workload management controller is configured to receive energy efficiency scores transmitted from the micro-data centers.

[0106] At S340, computing workloads are dynamically allocated. Dynamically allocating computing workloads among the plurality of micro-data centers is performed. In some embodiments, dynamically allocating computing workloads among the plurality of micro-data centers is performed in accordance with the received energy efficiency scores so as to maximize overall energy reuse for energy-optimized workloadmanagement. Maximizing overall energy reuse includes, in some embodiments, routing computing workloads to micro-data centers located in geographic regions having active heating or cooling demands.

[0107] In some embodiments, the central workload management controller is configured to consider temporal variations in local energy demand, including seasonal heating and cooling requirements, as well as meteorological events, when allocating workloads between the plurality of micro-data centers in accordance with the transmitted energy efficiency scores from the plurality of micro-data centers. For example, the central workload management controller is configured to allocate larger computing workloads to micro-data centers of premises located in geographically cooler conditions than to micro-data centers of premises located in geographically warmer conditions.

[0108] In some embodiments, the central workload management controller is configured to determine when a building proximate to at least one micro-data center of the plurality of micro-data centers requires heating.

[0109] According to this embodiment, when it is determined that a building proximate to the at least one micro-data center requires heating, the central workload management controller is configured to operate the at least one micro-data center. When it is determined that a building proximate to the at least one micro-data center does not require heating, the central workload management controller is configured to stop any operations of the at least one micro-data center. In further embodiments, the central workload management controller is configured to allocate workloads based on the energy efficiency scores of operating micro-data centers.

[0110] It should be understood that any reference to an element herein using a designation such as “first,” “second,” and so forth does not generally limit the quantity or order of those elements. Rather, these designations are generally used herein as a convenient method of distinguishing between two or more elements or instances of an element. Thus, a reference to the first and second elements does not mean that only two elements may be employed there or that the first element must precede the second element in some manner. Also, unless stated otherwise, a set of elements includes one or more elements.

[0111] As used herein, the phrase “at least one of’ followed by a listing of items means that any of the listed items can be utilized individually, or any combination of two or more of the listed items can be utilized. For example, if a system is described as including “at least one of A, B, and C,” the system can include A alone; B alone; C alone; A and B in combination; B and C in combination; A and C in combination; or A, B, and C in combination.

[0112] All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the principles of the disclosure and the concepts contributed by the inventor to further the art and are to be construed as being without limitation to such specifically recited examples and conditions.

Claims

CLAIMSWhat is claimed is:1 . A system for energy-optimized workload management, comprising: a plurality of micro-data centers, each configured to perform computing operations that generate heat as a byproduct of the computing operations, wherein the plurality of micro-data centers are geographically distributed; a thermal management module proximately disposed to each micro-data center, configured to utilize the generated heat for a non-computing application; a local monitoring module at each micro-data center of the plurality of microdata centers configured to determine energy efficiency scores of the plurality of microdata centers based on one or more localized energy demand conditions; and a central workload management controller communicatively coupled to the plurality of micro-data centers, the central workload management controller configured to: receive the energy efficiency scores from each of the plurality of micro-data centers and dynamically allocate computing workloads among the plurality of micro-data centers in accordance with the received energy efficiency scores so as to maximize at least overall energy efficiency across the system.

2. The system of claim 1 , wherein the computing operations include any one of: artificial intelligence (Al) processing, big data analytics, cryptocurrency mining, content delivery, and enterprise operations.

3. The system of claim 2, wherein Al processing includes training and inference of at least one of: machine learning models and language models.

4. The system of claim 1 , wherein the non-computing application comprises providing a heating function or a cooling function to an environment proximate to the micro-data center and to the micro-data center.

5. The system of claim 1 , wherein the non-computing application comprises providing cooling via a thermally-driven cooling system.

6. The system of claim 5, wherein the thermally-driven cooling system comprises an absorption chiller configured to utilize the generated heat to provide cooling.

7. The system of claim 6, wherein the absorption chiller is coupled to a heat pump integrated into the thermally-driven cooling system.

8. The system of claim 7, wherein the heat pump is mechanically-driven.

9. The system of claim 1 , wherein the energy efficiency scores represent a measure of at least one of: a potential capture, a potential reuse, and a potential renewability of the generated heat for a non-computing application, wherein the energy efficiency score is determined on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse.

10. The system of claim 9, wherein the energy efficiency score further comprises: a factor based on a source of electrical energy powering a respective micro-data center of the plurality of micro-data centers, wherein a renewable or green energy source contributes a higher score relative to a non-renewable energy source.

11. The system of claim 1 , wherein the central workload management controller is further configured to consider temporal variations in local energy demand, including seasonal heating and cooling requirements, and meteorological events, when allocating workloads.

12. The system of claim 1 , wherein the plurality of micro-data centers are configured as modular, distributed units operable within residential or commercial environments.

13. The system of claim 1 , wherein the system is further configured to: maximize the overall energy efficiency, energy reuse and energy cost saving of the system by routing computing workloads to micro-data centers located in geographic regions having active heating or cooling demands.

14. The system of claim 1 , wherein the plurality of micro-data centers are geographically distributed in areas with different temperature environments.

15. The system of claim 1 , wherein the central workload management controller is further configured to allocate larger computing workloads of the computing workloads to a micro-data center of the plurality of micro-data centers located in geographically cooler conditions than to a micro-data center of the plurality of micro-data centers located in geographically warmer conditions.

16. The system of claim 1 , wherein the central workload management controller is configured to: determine when a building proximate to at least one micro-data center of the plurality of micro-data centers requires heating; when it is determined that an environment proximate to the at least one microdata center requires heating, operate, by the central workload management controller, the at least one micro-data center; and when it is determined that an environment proximate to the at least one microdata center does not require heating, stop operations of the at least one micro-data center; and allocate workloads based on the energy efficiency scores of operating microdata centers of the plurality of micro-data centers.

17. The system of claim 1 , wherein the micro-data center of the plurality of microdata centers is integrated in any one of: an entertainment system, an HVAC system, asmart appliance, an industrial equipment, a telecommunications device, a heat-emitting transport system, and a consumer electronic device.

18. The system of claim 1 , wherein the micro-data center of the plurality of microdata centers is deployable in any one of: residential, commercial, and an enterprise environment.

19. The system of claim 1 , wherein the thermal management module includes: a desiccant system configured to absorb moisture from air upon thermal activation from the heat generated by the micro-data center.

20. A method for energy-optimized workload management, comprising: performing computing operations at a plurality of micro-data centers, each micro-data center generating heat as a byproduct of the computing operations, wherein the plurality of micro-data centers are geographically distributed; determining, at each micro-data center of the plurality of micro-data centers, an energy efficiency score of the micro-data center based on one or more localized energy demand conditions; transmitting the energy efficiency scores from the plurality of micro-data centers to a central workload management controller; and dynamically allocating computing workloads among the plurality of micro-data centers, by the central workload management controller, for a non-computing application in accordance with the determined energy efficiency scores so as to maximize at least overall energy efficiency of the plurality of micro-data centers.

21. The method of claim 20, wherein the computing operations include any one of: artificial intelligence (Al) processing, big data analytics, cryptocurrency mining, content delivery, and enterprise operations.

22. The method of claim 21 , wherein Al processing includes training and inference of at least one of: machine learning models and language models.

23. The method of claim 20, wherein the non-computing application comprises: providing a heating function or a cooling function to an environment proximate to at least one of the plurality of micro-data centers and to at least one of the plurality of micro-data centers.

24. The method of claim 20, wherein the non-computing application comprises: providing cooling via a thermally-driven cooling system.

25. The method of claim 24, wherein the thermally-driven cooling system comprises an absorption chiller configured to utilize the generated heat to provide cooling.

26. The method of claim 25, wherein the absorption chiller is coupled to a heat pump integrated into the thermally-driven cooling system.

27. The method of claim 26, wherein the heat pump is mechanically-driven.

28. The method of claim 20, wherein the energy efficiency score represents a measure of potential capture, potential reuse, and potential renewability of the generated heat for a non-computing application, and wherein determining the energy efficiency score further comprises: assigning a value on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse.

29. The method of claim 20, further comprising: adjusting the allocated computing workloads based on temporal variations in local energy demand, including seasonal heating and cooling requirements, and meteorological events.

30. The method of claim 20, further comprising: incorporating a factor based on a source of electrical energy powering a respective micro-data center, wherein arenewable or green energy source contributes a higher energy efficiency score relative to a non-renewable energy source.

31. The method of claim 20, wherein maximizing the overall energy efficiency of the plurality of micro-data centers further comprises: maximizing energy reuse and energy cost saving by routing computing workloads to micro-data centers located in geographic regions having active heating or cooling demands.

32. The method of claim 20, wherein the plurality of micro-data centers are geographically distributed in areas with different temperature environments.

33. The method of claim 20, wherein dynamically allocating computing workloads among the plurality of micro-data centers, by the central workload management controller, further comprises: allocating larger computing workloads of the computing workloads to a micro-data center of the plurality of micro-data centers located in geographically cooler conditions than to a micro-data center of the plurality of microdata centers located in geographically warmer conditions.

34. The method of claim 20, further comprising: determining when a building proximate to at least one micro-data center of the plurality of micro-data centers requires heating; when it is determined that an environment proximate to the at least one microdata center requires heating, operating, by the central workload management controller, the at least one micro-data center; when it is determined that an environment proximate to the at least one microdata center does not require heating, stopping operations of the at least one micro-data center; and allocating workloads based on the energy efficiency scores of operating microdata centers of the plurality of micro-data centers.

35. The method of claim 20, wherein the micro-data center of the plurality of microdata centers is integrated in any one of: an entertainment system, an HVAC system, a smart appliance, an industrial equipment, a telecommunications device, a heat-emitting transport system, and a consumer electronic device.

36. The method of claim 20, wherein the micro-data center of the plurality of microdata centers is deployable in any one of: residential, commercial, and an enterprise environment.

37. The method of claim 24, wherein the thermally-driven cooling system includes: a desiccant system configured to absorb moisture from air upon thermal activation from the heat generated by the micro-data center.

38. A micro-data center, comprising: a computing module configured to perform data computing operations and to generate heat as a byproduct of the data computing operations; a thermal management module integrated with the computing module and configured to utilize at least a portion of the generated heat to provide a heating or cooling function to an environment proximate to the micro-data center and to the microdate center; and a communications interface configured to enable the micro-data center to exchange data with external computing networks, wherein the micro-data center is configured as a self-contained system such that both computing services and heating or cooling services are provided from a single integrated product.

39. The micro-data center of claim 38, wherein the data computing operations include any one of: artificial intelligence (Al) processing, big data analytics, cryptocurrency mining, content delivery and enterprise operations.

40. The micro-data center of claim 39, wherein Al processing includes training and inference of at least one of: machine learning models and language models.

41. The micro-data center of claim 40, wherein the computing module includes at least one graphics processing unit (GPU).

42. The micro-data center of claim 41 , wherein the computing module further comprises: a central processing unit (CPU), a memory, and a storage device.

43. The micro-data center of claim 40, wherein the thermal management module comprises a cooling module.

44. The micro-data center of claim 43, wherein the cooling module comprises a thermally-driven cooling system.

45. The micro-data center of claim 44, wherein the thermally-driven cooling system comprises an absorption chiller configured to utilize the generated heat to provide cooling.

46. The micro-data center of claim 44, wherein the thermally-driven cooling system comprises a desiccant system configured to utilize the generated heat to provide cooling.

47. The micro-data center of claim 45, wherein the thermally-driven cooling system further comprises a desiccant system.

48. The micro-data center of claim 45, wherein the absorption chiller is coupled to a heat pump integrated into the thermally-driven cooling system.

49. The micro-data center of claim 48, wherein the heat pump is mechanically- driven.

50. The micro-data center of claim 40, further comprising a central workload management controller configured to compute an energy efficiency score of the microdata center based on one or more localized energy demand conditions, the energy efficiency score representing a measure of potential capture, potential reuse, and potential renewability of the generated heat for a non-computing application.

51. The micro-data center of claim 50, wherein the energy efficiency score is determined on a normalized scale between zero and ten, where zero represents no recoverable energy reuse and ten represents complete recoverable energy reuse.

52. The micro-data center of claim 51 , wherein the energy efficiency score further comprises a factor based on a source of electrical energy powering the micro-data center, wherein a renewable or green energy source contributes a higher score relative to a non-renewable energy source.

53. The micro-data center of claim 50, wherein the central workload management controller is configured to dynamically allocate computing workloads among a plurality of micro-data centers in accordance with the computed energy efficiency scores so as to maximize overall energy efficiency, energy reuse, and energy cost saving across the plurality of micro-data centers.

54. The micro-data center of claim 50, wherein the central workload management controller is configured to: determine when a building proximate to at least one micro-data center of a plurality of micro-data centers requires heating; when it is determined that an environment proximate to the at least one microdata center requires heating, operate, by the central workload management controller, the at least one micro-data center; when it is determined that an environment proximate to the at least one microdata center does not require heating, stop operations of the at least one micro-data center; andallocate workloads based on the energy efficiency scores of operating microdata centers of the plurality of micro-data centers.

55. The micro-data center of claim 40, wherein the micro-data center is integrated in any one of: an entertainment system, an HVAC system, a smart appliance, an industrial equipment, a telecommunications device, a heat-emitting transport system, and a consumer electronic device.

56. The micro-data center of claim 40, wherein the micro-data center is deployable in any one of: residential, commercial, or an enterprise environment.

57. A system for thermal energy reuse in a computing environment, comprising: a computing device configured to generate heat during operation; a heat transfer medium configured to absorb thermal energy from the computing device; a transport conduit thermally coupled to the computing device and configured to convey the thermal energy carried by the heat transfer medium from a first location adjacent to the computing device to a second location remote from the computing device; and a secondary energy utilization system located at the second location and configured to receive the thermal energy carried by the heat transfer medium and utilize the thermal energy to perform a non-computational process.

58. The system of claim 57, wherein the secondary energy utilization system comprises a thermally-activated cooling mechanism configured to remove heat from the computing device, such that the thermal energy generated by the computing device is reused by the thermally-activated cooling mechanism to facilitate active cooling of the computing device itself.

59. The system of claim 57, wherein the heat transfer medium further comprises: a liquid desiccant or another thermally conductive liquid.

60. The system of claim 57, wherein the non-computational process comprises at least one of environmental cooling, dehumidification, or mechanical actuation.

Images