The present disclosure generally relates to information handling systems (IHS), and more particular to a rack-configured IHS, having a liquid cooling subsystem and liquid-cooled nodes. Still more particularly, the disclosure is related to a liquid cooling system having a modular rack liquid distribution network of conduits.
As the value and use of information continue to increase, individuals and businesses seek additional ways to process and store information. One option available to users is Information Handling Systems (IHSs). An IHS generally processes, compiles, stores, and/or communicates information or data for business, personal, or other purposes, thereby allowing users to take advantage of the value of the information. Because technology and information handling needs and requirements vary between different users or applications, IHSs may also vary regarding what information is handled, how the information is handled, how much information is processed, stored, or communicated, and how quickly and efficiently the information may be processed, stored, or communicated. The variations in IHSs allow for IHSs to be general or configured for a specific user or specific use such as financial transaction processing, airline reservations, enterprise data storage, or global communications. In addition, IHSs may include a variety of hardware and software components that may be configured to process, store, and communicate information and may include one or more computer systems, data storage systems, and networking systems.
For implementations requiring a large amount of processing capability, a rack-configured (or rack) IHS (RIHS) can be provided. The RIHS includes a physical rack, within which is inserted a plurality of functional nodes, such as server (or processing) nodes/modules, storage nodes, and power supply nodes. These nodes, and particularly the server nodes, typically include processors and other functional components that dissipate heat when operating and/or when connected to a power supply. Efficient removal of the heat being generated by these components is required to maintain the operational integrity of the RIHS. Traditional heat removal systems include use of air movers, such as fans, to convectionally transfer the heat from inside of the RIHS to outside the RIHS. More recently, some RIHS have been designed to enable submersion of the server modules and/or the heat generating components in a tank of cooling liquid to effect cooling via absorption of the heat by the surrounding immersion liquid.
The amount of processing capacity and storage capacity per node and/or per rack continues to increase, providing greater heat dissipation per node and requiring more directed cooling solutions. Thus, there is a continuing need for further innovations to provide directed cooling for the individual heat generating components, both at the individual node level, as well as at the larger rack level. When designing the cooling subsystem, consideration must also be given to the different form factors of IT nodes and rack heights of the RIHS, and the ability to effectively control cooling discretely (at device or node level) and generally across the overall RIHS.
As liquid cooling improves in efficiencies and performance, data center solutions continue to focus on implementing liquid cooling at the rack level. Recently, localized liquid solutions (CPU/GPU cold plates) have been successful in removing most of the heat from these components within a server and into the facility cooling loop through direct fluid-to-fluid heat exchangers (server cooling loop to facility cooling loop) within the rack, but this method does not provide cooling to auxiliary components (such as storage devices (HDDs, memory), or critical secondary ICT equipment, such as top of the rack switch, network switches, battery backup units, or Power Supply Units (PSUs).
The illustrative embodiments of the present disclosure provides a Direct-Interface Liquid-Cooled (DL) Rack Information Handling System (RIHS), a direct-interface liquid cooling subsystem, and a method for modularly providing liquid cooling to information technology (IT) nodes within a RIHS. The IT nodes are liquid cooled (LC) nodes that contain heat-generating functional components, and the nodes can have different form factors and different cooling requirements. In one or more embodiments, the RIHS includes a rack having chassis-receiving bays. Each chassis is received into a respective chassis-receiving bay of the rack. The chassis can be a node enclosure or can be a block chassis that receives more than one node enclosure. Each LC node is configured with a system of conduits to receive direct injection of cooling liquid to regulate the ambient temperature of the node and provide cooling to the components inside the node by absorbing and removing heat generated by the heat-generating functional components. The cooling subsystem has a liquid rail formed by more than one node-to-node, Modular Liquid Distribution (MLD) conduits connecting between LC nodes provisioned in the rack. The MLD conduits comprise first and second terminal connections attached on opposite ends of a central conduit that is rack-unit dimensioned to seal to and enable fluid transfer between a port of a selected LC node and a port of an adjacent LC node.
According to at least one aspect of the present disclosure, a liquid cooling subsystem of a RIHS includes a liquid rail formed by more than one node-to-node, MLD conduits connecting between more than one LC node provisioned in a rack. Each LC node includes a chassis received in a respective chassis-receiving bay of the rack and each LC node contains one or more heat-generating functional components. Each LC node is configured with a system of conduits to receive direct injection of cooling liquid to regulate the ambient temperature of the node and provide cooling to the components inside the node by absorbing and removing heat generated by the heat-generating functional components.
According to at least one aspect of the present disclosure, a method is provided of assembling a DL RIHS that includes mounting a plurality of block liquid manifolds at a back of the rack, where each block liquid manifold has at least one rail supply port and a rail return port on an outside facing side of the block liquid manifold that respectively communicate with at least one block supply port and a block return port on an inside facing side of the block liquid manifold. The method includes inserting more than one LC nodes in receiving bays of the rack corresponding to locations of the mounted block liquid manifolds, with the supply ports and the return ports of the LC nodes and an inside facing portion of the corresponding block liquid manifold linearly aligned. The method includes sealing, for fluid transfer, the ports of a selected one or more LC nodes to the lineally aligned block supply and return ports of the inside facing portion of the block liquid manifold. The method includes attaching node-to-node interconnecting Modular Liquid Distribution (MLD) conduits between a selected pair of adjacent rail supply ports and adjacent rail return ports of the exterior face of the block liquid manifolds to form a supply conduit and a return conduit, respectively, of a scalable liquid rail. The method further includes attaching bypass conduits across the nodes to allow for continued flow of cooling liquid to the remaining nodes in the RIHS following removal of one or more of the MLD conduits serving specific nodes. The assembly enables liquid cooling to the functional components inside the LC nodes. Liquid cooling is provided to the functional components inside the LC nodes via liquid absorption of heat generated by the heat-generating functional components, followed by liquid transfer of the absorbed heat away from the inside of the node.
The above presents a general summary of several aspects of the disclosure in order to provide a basic understanding of at least some aspects of the disclosure. The above summary contains simplifications, generalizations and omissions of detail and is not intended as a comprehensive description of the claimed subject matter but, rather, is intended to provide a brief overview of some of the functionality associated therewith. The summary is not intended to delineate the scope of the claims, and the summary merely presents some concepts of the disclosure in a general form as a prelude to the more detailed description that follows. Other systems, methods, functionality, features and advantages of the claimed subject matter will be or will become apparent to one with skill in the art upon examination of the following figures and detailed written description.
The description of the illustrative embodiments can be read in conjunction with the accompanying figures. It will be appreciated that for simplicity and clarity of illustration, elements illustrated in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements are exaggerated relative to other elements. Embodiments incorporating teachings of the present disclosure are shown and described with respect to the figures presented herein, in which:
The present disclosure generally provides a Direct-Interface Liquid-Cooled (DL) Rack Information Handling System (RIHS) providing liquid cooled (LC) information technology (IT) nodes containing heat-generating functional components and which are cooled at least in part by a liquid cooling subsystem. The RIHS includes a rack configured with chassis-receiving bays in which is received a respective chassis of one of the LC nodes. Each LC node is configured with a system of conduits to receive direct injection of cooling liquid to regulate the ambient temperature of the node. Additionally, each LC node, configured with a system of conduits, provides cooling to the components inside the node by conductively absorbing, via the cooling liquid, heat generated by the heat-generating functional components. The absorbed heat is removed (or transferred away) from within the node to outside of the node and/or the RIHS. The cooling subsystem has a liquid rail formed by more than one node-to-node, Modular Liquid Distribution (MLD) conduit, which include first and second terminal connections attached on opposite ends of a central conduit. The MLD conduits are rack-unit dimensioned to seal to and enable fluid transfer between a port of a selected LC node and a port of an adjacent LC node.
In the following detailed description of exemplary embodiments of the disclosure, specific exemplary embodiments in which the disclosure may be practiced are described in sufficient detail to enable those skilled in the art to practice the disclosed embodiments. For example, specific details such as specific method orders, structures, elements, and connections have been presented herein. However, it is to be understood that the specific details presented need not be utilized to practice embodiments of the present disclosure. It is also to be understood that other embodiments may be utilized and that logical, architectural, programmatic, mechanical, electrical and other changes may be made without departing from general scope of the disclosure. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present disclosure is defined by the appended claims and equivalents thereof.
References within the specification to “one embodiment,” “an embodiment,” “embodiments”, or “one or more embodiments” are intended to indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. The appearance of such phrases in various places within the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Further, various features are described which may be exhibited by some embodiments and not by others. Similarly, various requirements are described which may be requirements for some embodiments but not other embodiments.
It is understood that the use of specific component, device and/or parameter names and/or corresponding acronyms thereof, such as those of the executing utility, logic, and/or firmware described herein, are for example only and not meant to imply any limitations on the described embodiments. The embodiments may thus be described with different nomenclature and/or terminology utilized to describe the components, devices, parameters, methods and/or functions herein, without limitation. References to any specific protocol or proprietary name in describing one or more elements, features or concepts of the embodiments are provided solely as examples of one implementation, and such references do not limit the extension of the claimed embodiments to embodiments in which different element, feature, protocol, or concept names are utilized. Thus, each term utilized herein is to be given its broadest interpretation given the context in which that terms is utilized.
As utilized herein, the term “rack-configured” (as in RIHS) generally refers to the configuration of a large scale sever system within a physical rack having multiple chassis receiving rails for receiving specific sizes of information technology (IT) nodes, such as server modules, storage modules, and power modules. The term node generally refers to each separate unit inserted into a 1U or other height rack space within the rack. In one embodiment, operational characteristics of the various IT nodes can be collectively controlled by a single rack-level controller. However, in the illustrated embodiments, multiple nodes can be arranged into blocks, with each block having a separate block-level controller that is communicatively connected to the rack-level controller.
For purposes of this disclosure, an information handling system (defined at the individual server level) may include any instrumentality or aggregate of instrumentalities operable to compute, classify, process, transmit, receive, retrieve, originate, switch, store, display, manifest, detect, record, reproduce, handle, or utilize any form of information, intelligence, or data for business, scientific, control, or other purposes. For example, an information handling system may be a personal computer, a network storage device, or any other suitable device and may vary in size, shape, performance, functionality, and price. The information handling system may include random access memory (RAM), one or more processing resources such as a central processing unit (CPU) or hardware or software control logic, ROM, and/or other types of nonvolatile memory. Additional components of the information handling system may include one or more disk drives, one or more network ports for communication with external devices as well as various input and output (I/O) devices, such as a keyboard, a mouse, and a video display. The information handling system may also include one or more buses operable to transmit communications between the various hardware components.
As illustrated by the figures and described herein, multiple processing servers or server IHSs (referred to herein as server nodes) can be included within the single RIHS. Certain aspect of the disclosure then relate to the specific LC (sever or other) nodes and the functionality associated with these individual nodes or block-level groupings of nodes, while other aspects more generally relate to the overall DL RIHS containing all of the LC nodes.
As one design detail/aspect for the present innovation, consideration is given to the fact that extreme variations can exist in server/power/network topology configurations within an IT rack. In addition to dimension variations, the thermal requirements for heat-generating functional components for power, control, storage and server nodes can be very different between types or vary according to usage. These variations drive corresponding extreme diversity in port placement, fitting size requirements, mounting locations, and manifold capacity for a liquid cooling subsystem. Further, a chassis of each node is typically densely provisioned. Lack of space thus exists to mount a discrete water distribution manifold in high-power IT racks. The present disclosure addresses and overcomes the challenges with distributing liquid cooling fluids throughout an IT rack having nodes with a large number of variations in distribution components.
The disclosure also includes the additional consideration that in addition to cooling the primary heat generating components of the rack, such as the processor, what is needed is a way to allow for cooling of secondary equipment within the rack, as well as auxiliary components that would further support utilizing the advantages of a fluid-to-fluid heat exchanger methodology. Additionally, the present disclosure provides a modular approach to utilizing an air-to-liquid heat exchanger with quick connection and scalability to allow the solution to be scalable in both 1U and 2U increments.
The present innovation is not limited to any specific number or configuration of nodes 102 or blocks 112 in a rack 104. According to one aspect, nodes 102 can be of different physical heights of form factors (e.g., 1U, 1.5U, 2U), and the described features can also be applied to nodes 102 having different widths and depths (into the rack), with some extensions made and/or lateral modifications to the placement of cooling subsystem conduits, as needed to accommodate the different physical dimensions. As a specific example, node 102i is depicted as having a larger node enclosure 107′ (with corresponding different dimensions of heat-generating functional components 106′) of a different number of rack units in physical height (e.g., 2U) that differs from the heights (e.g., 1U) of the other nodes 102a-102h and 102j-102k. RIHS 100 can include blocks 112 or nodes 102 selectably of a range of discrete rack units. Also, different types of IT components can be provided within each node 102, with each node possibly performing different functions within RIHS 100. Thus, for example, a given node 102 may include one of a server module, a power module, a control module, or a storage module. In a simplest configuration, the nodes 102 can be individual nodes operating independent of each other, with the RIHS 100 including at least one rack-level controller (RC) 116 for controlling operational conditions within the RIHS 100, such as temperature, power consumption, communication, and the like. Each node 102 is then equipped with a node-level controller (NC) 118 that communicates with the rack-level controller 116 to provide localized control of the operational conditions of the node 102. In the more standard configuration of a DL RIHS 100, and in line with the described embodiments, RIHS 100 also includes block-level controllers (BCs) 114, communicatively coupled to the rack-level controller 116 and performing block-level control functions for the LC nodes within the specific block. In this configuration, the nodes 102 are arranged into blocks 112, with each block 112 having one or more nodes 102 and a corresponding block-level controller 114. Note the blocks do not necessarily include the same number of nodes, and a block can include a single node, in some implementations.
A Direct-Interface Liquid Cooling (DL) subsystem (generally shown as being within the RIHS and labelled herein as 120) provides liquid cooling to heat-generating functional components 106 via a liquid rail 124 under the control of the rack-level controller 116, block-level controllers 114, and/or node-level controllers 118, in some embodiments. Rack-level controller 116 controls a supply valve 126, such as a solenoid valve, to allow cooling liquid, such as water, to be received from a facility supply 128. The cooling liquid is received from facility supply 128 and is passed through liquid filtration node 102l before being passed through supply conduit 130 of liquid rail 124. Each block 112a, 112b receives a dynamically controlled amount of the cooling liquid via block-level dynamic control valve 132, such as a proportional valve. Return flow from each block 112a, 112b can be protected from backflow by a block check valve 133. The individual needs of the respective nodes 102a-102d of block 112a can be dynamically provided by respective node-level dynamic control valves 134, controlled by the block-level controller 114, which control can, in some embodiments, be facilitated by the node-level controllers 118. In addition to allocating cooling liquid in accordance with cooling requirements (which can be optimized for considerations such as performance and economy), each of the supply valve 126 and/or dynamic control valves 132, 134 can be individually closed to mitigate a leak. A check valve 136 is provided between each node 102a-102j and a return conduit 138 of the liquid rail 124 to prevent a backflow into the nodes 102a-102j. The return conduit 138 returns the cooling liquid to a facility return 140.
To support the temperature control aspects of the overall system, RIHS 100 includes temperature sensors 101 that are each located within or proximate to each node 102a-102j, with each gauge 101 connected to the node-level controller 118 and/or the corresponding block-level controller 114. Temperature sensors 101 operate in a feedback control loop of the liquid cooling system 122 to control the amount of liquid flow required to cool the nodes 102a-102j. In one or more embodiments, the rack-level controller 116 can coordinate performance constraints to block-level controllers 114 and/or node-level controllers 118 that limit an amount of heat generated by the heat-generating functional components 106 to match a heat capacity of the flow of cooling liquid in DL subsystem 122. Alternatively or in addition, the rack-level controller 116 can coordinate cooling levels to block-level controllers 114 and/or node-level controllers 118 that in turn control the dynamic control valves 132, 134 for absorption and transfer of the heat generated by the heat-generating functional components 106 by the DL subsystem 122. In one or more embodiments, support controllers such as a Rack Liquid Infrastructure Controller (RLIC) 142 can perform management and operational testing of DL subsystem 122. RLIC 142 can monitor pressure sensors 144 and liquid sensors 146 to detect a leak, to validate operation of a namic control valves 132, 134 or shut-off valves such as supply valve 126. RLIC 142 can perform close-loop control of specific flow rates within the RIHS 100.
Illustrated in
At the top position of RIHS 800, a block chassis 810 is received in a block chassis receiving bay 870a of rack 804. Within block chassis 810, a first node 802a received in a first node receiving bay 809a of the rack 804 has a vertical height of one rack unit (1U). A rack unit, U or RU as a unit of measure, describes the height of electronic equipment designed to mount in a 19-inch rack or a 13-inch rack. The 19 inches (482.60 mm) or 13 inches (584.20 mm) dimension reflects the horizontal lateral width of the equipment mounting-frame in the rack including the frame; the width of the equipment that can be mounted inside the rack is less. According to current convention, one rack unit is 1.75 inches (44.45 mm) high. A second node 802b received in a second node receiving bay 809b of the rack 104 (of
MLD conduits 878a of 1U can be used to connect nodes of 1U vertical spacing. Because of the additional 1U separation of LC nodes 802c and 802e by inclusion of infrastructure node 802d, MLD conduit 878b between the third and fifth nodes 802c-802d is dimension 2U to accommodate the increased spacing. MLD conduits 878a-878b can thus support different heights (1U to NU) of IT components.
Each MLD conduit 878a-878b includes first and second terminal connections 883, 884 attached on opposite ends of central conduit 885 that is rack-unit dimensioned to seal to a port of LC node 802 and enable fluid transfer between a port of a selected LC node 802 and a port of an adjacent LC node 802. In
Liquid cooling subsystem 822 includes a liquid infrastructure manager controller (LIMC) 886 which is communicatively coupled to block liquid controllers (BLCs) 887 to collectively control the amount of cooling liquid that flows through the RIHS 800 and ultimately through each of the nodes 802 in order to effect a desired amount of liquid cooling at the component level, node level, block level, and rack level. For clarity, LIMC 886 and BLCs 887 are depicted as separate components. In one or more embodiments, the liquid control features of the LIMC 886 and BLCs 887 can be incorporated into one or more of the rack-level controller 816, block-level controllers 820, and the node-level controllers 818. As illustrated in
As another aspect, the present disclosure provides a modular approach to utilizing air-to-liquid heat exchanger 888 with quick connection and is scalable in both 1U and 2U increments. In one or more embodiments, DL cooling subsystem 822 can include a plurality of air-to-liquid (or liquid-to-air) heat exchangers 888 that facilitate the release of some of the heat absorbed by the exhaust liquid to the surrounding atmosphere around the RIHS 100 (of
RIHS 800 can include variations in LC node 802 that still maintain uniformity in interconnections along liquid rail 824 formed by a chassis-to-chassis modular interconnect system of MLD conduits 878a-878b. With this scalability feature accomplished using MLD conduits 878a-878b, cooling subsystem 822 of the RIHS 800 allows each block chassis 810 to be a section of a scalable manifold, referred herein as liquid rail 824, eliminating the need for a rack manifold. The scalability of liquid rail 824 enables flexible configurations to include various permutations of server and switch gear within the same rack (rack 804). MLD conduits 878a-878b can comprise standardized hoses with sealable (water tight) end connectors. Thus, the rack liquid flow network can encompass 1 to N IT chassis without impacting rack topology, space constraints, and without requiring unique rack manifolds. Additionally, according to one aspect, the MLD conduits are arranged in a pseudo daisy chain modular configuration, which allows for unplugging of one MLD conduit from one rack level without affecting liquid flow to and cooling of other rack levels.
The system of conduits extending from node intake valve 834 into each LC node 802 enables each LC node 802 to engage to block liquid manifold 889. Block chassis 810 or node enclosure 808 of each LC node 102 provides the intake and exhaust conduit connections to engage to respective terminals of MLD conduits 878a-878b within the MLD network provided by liquid rail 824. For example, where nodes 802 are designed as sleds, node enclosure 808 would be a sled tray, and each block would then include more than one sled tray received into block chassis 810, forming the extensions of block liquid manifold 889. Alternatively, the node enclosure 808 can be a single node chassis such as one of nodes 802c-802f.
Supply and return bypass tubes 890, 891 of each block liquid manifold 889 are connected by MLD conduits 878a-878b to form supply rail conduit 830 and return rail conduit 838. Due to constraints in the spacing within the figure, the tubing that extends from supply and return bypass tubes 890, 891 are not shown, and the valves are shown as if connected directly to the bypass.
In an exemplary embodiment, the scalable rack manifold provided by liquid rail 824 is formed in part by MLD conduits 878a-878b that run vertically in the back of the RIHS 800 with quick disconnects on the front and rear face of block liquid manifold 889 that allows for IT/infrastructure equipment respectively to be plugged into both front and back sides of the block liquid manifold 889. For example, LC nodes 802, such as server modules, can plug into the front side and fan modules 882 can plug onto the back side of block liquid manifold 889. This also allows for other liquid cooled devices such as LC Power Distribution Units (PDUs) to be plugged into the cooling liquid supply rail conduit 830 and return rail conduit 838 of liquid rail 824. Thereby, a rack hot pluggable cooling interface is created for any rack-mounted equipment.
Cooling subsystem 822 can support an embedded liquid-to-liquid heat exchanger manifold 842, such as in LC node 802c. Node liquid-to-liquid heat exchangers are provided for rejecting heat from one fluid source to a secondary source. One aspect of the present disclosure solves the problems that many shared-infrastructure IT systems (e.g., blade chassis) do not have adequate space to accommodate a liquid-to-liquid heat exchanger. Unlike with generally-known systems that rely upon liquid heat transfer having to exchange heat with an external liquid-to-liquid heat exchanger, the present disclosure enables on-rack liquid-to-liquid heat exchanger that does not require any of the vertical chassis space. Additionally, the present disclosure provides these benefits without requiring a central distribution unit (CDU), which takes up datacenter floor space. One aspect of the present disclosure provides embedded heat exchanger manifold 842 having a common heat transfer plate and a shared bulk header to create a combined liquid distribution manifold that includes a secondary liquid coolant for absorbing heat through the shared bulk header. In particular, the combined embedded heat exchanger manifold 842 rejects heat within shared node enclosure 808 such as node 802c to a secondary liquid coolant. Internal node supply 844 and return conduits 846 of a manifold built on top of a heat exchanger core allow heat transport within manifold 842. In one embodiment, closed system pump 898 can use a first coolant to cool a high thermal energy generating functional component such as a CPU or voltage regulator.
Additionally, the liquid cooling subsystem 822 also includes a filtration system or unit 871, which prevents chemical impurities and particulates from clogging or otherwise damaging the conduits as the fluid passes through the network of conduits. According to one aspect of the disclosure, liquid cooling subsystem 822 provides RFU 871 in fluid connection with the intake pipes from facility supply 828. In at least one embodiment, RFU 871 includes a sequenced arrangement of liquid filters within a full-sized sled that can be removably inserted by an end user into one of the receiving slots of rack 804. In one embodiment, the RFU 871 is located on an infrastructure sled having rack-level controllers and other rack-level functional components. In at least one embodiment, the entirety of the sled is filed with components associated with RFU 871. Thus, it is appreciated that the RFU 871 may occupy the entire area of one vertical slot/position within the chassis. Alternate locations of the RFU 871 can also be provided, in different embodiments, with an ideal location presenting the intake port of the RFU 871 in close proximity to a connection to facility supply 828 to directly receive the facility supply 828 prior to the liquid being passed into the remainder of the conduits of the liquid cooling subsystem 822. It is appreciated that if the system was capable of completing all heat exchange within the rack, then sealing the rack would be feasible and would reduce and/or remove any requirements for filtration and/or allocation of rack space for RFU 871.
Liquid cooled compute systems use the high heat transport capacity of water. However, the disclosure recognizes and addresses the fact that with liquid introduced into an electronic enclosure, there is a potential for leaks that can cause catastrophic system failure. Also, in some instances, a leak can create an electronic short with a resulting exothermal reaction causing permanent damage to the DL RIHS 800. To mitigate such risks, as one design feature, node-level carrier 893 can include a trench/gutter system for use as liquid containment structure 894. In one embodiment, the gutter system can also incorporate an absorbent material that can accumulate sufficient amounts of liquid from small leaks to enable external sensing of the leak. Advantageously, the carrier 893 can also be thermally conductive to serve as a heat sink for components such as storage devices 895. In one embodiment, another leak detection solution that can be incorporated into the LC node 802 involves use of a solenoid to create an event when additional current is applied, due to water pooling around the solenoid. Barriers on carrier 893 can be specifically designed to contain a liquid leak and assist in funneling the liquid through the gutter system. Liquid rail 824 can also be provided with leak containment and detection. In one or more embodiments, removable pipe covers 876 are sized to be mounted around respective MLD conduits 878a-878b and can include liquid sensors 897 for automatic alerts and shutdown measures.
In one or more embodiments, DL RIHS 800 further incorporates a node-level liquid containment structure 890 with a cascading drain runoff tubing network 896 to a rack-level cascading liquid containment structure 894. In one or more embodiments, the DL RIHS 800 further incorporates leak detection response such as partial or complete automated emergency shutdown. Liquid sensors (LS) 897 at various cascade levels can identify affected portions of DL RIHS 800. Containment and automatic shutoff can address the risks associated with a leak developing in the DL cooling system 822.
The cooling subsystem 920 includes block liquid manifolds 989 mountable at a back side of the rack 904. Each block liquid manifold has at least one rail supply port 917 and at least one rail return port 919 on an outside facing side of the block liquid manifold 989. The at least one rail supply port 917 and the at least one rail return port 919 respectively communicate with at least one block supply port 921 and a block return port 923 on an inside facing side of the block liquid manifold 989. LC nodes 902 are insertable in receiving bays 970 of rack 904 corresponding to locations of the mounted block liquid manifolds 989. Block supply ports 921 and block return ports 923 of the LC nodes 902 and an inside facing portion of the corresponding block liquid manifold 989 are linearly aligned. The linear alignment enables direct sealing, for fluid transfer, of the lineally aligned inside manifold supply ports 925 and return ports 927 to the inside facing portion of the block liquid manifold 989. In one or more embodiments, block supply port 921 sealed to the internal manifold supply port 925 communicates via supply bypass tube 990 to two rail supply ports 917. Block return port 923 sealed to internal manifold return port 927 communicates via return bypass tube 991 of the respective block liquid manifold 989 to two rail return ports 919. Fan modules 982 mounted respectively onto back of block liquid manifold 989 have apertures to expose rail supply 917 and return ports 919. Additionally, fan modules 982 draw hot air 999 from LC nodes 902 through an air-liquid heat exchanger 988 in block liquid manifold 989.
In one or more embodiments, supply liquid conduit 992a is attached for fluid transfer between facility supply 928 and rail supply port 917 of block liquid manifold 989 of RIHS 900. A return liquid conduit 992b can be attached for fluid transfer between rail return port 919 of block liquid manifold 989 to facility return 940.
In one or more embodiments, RIHS 900 can provide hot-pluggable server-level liquid cooling, an integrated leak collection and detection trough, and an automatic emergency shut-off circuit. At a block level, RIHS 900 can provide embedded air-to-liquid heat exchange, and dynamic liquid flow control. At a rack level, RIHS 900 can provide facility-direct coolant delivery, a scalable rack fluid network, a rack filtration unit, and automated rack flow balancing, and a service mode.
According to one embodiment, liquid rail 924 includes a series of secondary conduits, such as supply divert conduit 997 and return divert conduit 998 that provides a by-pass fluid path for each of MLD conduits 978. In operation, divert conduit 997 allows for the removal of corresponding MLD conduit 978, thus removing the flow of cooling liquid to the particular block of nodes, without interrupting the flow of cooling liquid to the other surrounding blocks of computer gear. For example, a particular MLD conduit 978 can be replaced due to a leak. For another example, a block liquid manifold 989 can be replaced. The inclusion of divert conduits 997, 998 thus enables rapid servicing and maintenance of block liquid manifold 989 and/or nodes within block chassis without having to reconfigure the MLD conduits 978. In addition, the RIHS 900 can continue operating as cooling liquid continues to be provided to the remainder of the blocks that are plugged into the liquid rail. Re-insertion of the MLD conduit 978 then reconnects the flow of cooling liquid to the block for normal cooling operations, and shuts off the diverted flow of cooling liquid. In an exemplary embodiment, the MLD conduits 978 provide a quick disconnect feature that interrupts flow when not fully engaged to a respective port 917, 919, 921, 923. Disconnection of an MLD conduit 978 interrupts flow in a primary portion of the liquid rail 924 for either supply or return, shifting flow through one or more divert conduits 997 to provide cooling liquid to the other block liquid manifolds 989. In one or more embodiments, a manual or active shutoff valve can interrupt flow on either or both of the primary or divert portions of the liquid rail 924.
In the above described flow charts of
One or more of the embodiments of the disclosure described can be implementable, at least in part, using a software-controlled programmable processing device, such as a microprocessor, digital signal processor or other processing device, data processing apparatus or system. Thus, it is appreciated that a computer program for configuring a programmable device, apparatus or system to implement the foregoing described methods is envisaged as an aspect of the present disclosure. The computer program may be embodied as source code or undergo compilation for implementation on a processing device, apparatus, or system. Suitably, the computer program is stored on a carrier device in machine or device readable form, for example in solid-state memory, magnetic memory such as disk or tape, optically or magneto-optically readable memory such as compact disk or digital versatile disk, flash memory, etc. The processing device, apparatus or system utilizes the program or a part thereof to configure the processing device, apparatus, or system for operation.
While the disclosure has been described with reference to exemplary embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the disclosure. In addition, many modifications may be made to adapt a particular system, device or component thereof to the teachings of the disclosure without departing from the essential scope thereof. Therefore, it is intended that the disclosure not be limited to the particular embodiments disclosed for carrying out this disclosure, but that the disclosure will include all embodiments falling within the scope of the appended claims. Moreover, the use of the terms first, second, etc. do not denote any order or importance, but rather the terms first, second, etc. are used to distinguish one element from another.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The description of the present disclosure has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the disclosure in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope of the disclosure. The described embodiments were chosen and described in order to best explain the principles of the disclosure and the practical application, and to enable others of ordinary skill in the art to understand the disclosure for various embodiments with various modifications as are suited to the particular use contemplated.
The present invention claims priority from Provisional Application Ser. No. 62/270,563, with filing/priority date of Dec. 21, 2015. The entire content of the provisional application is incorporated herein by reference.
Number | Name | Date | Kind |
---|---|---|---|
6775137 | Chu et al. | Aug 2004 | B2 |
7013955 | Phillips et al. | Mar 2006 | B2 |
7403384 | Pflueger | Jul 2008 | B2 |
8405975 | Helberg et al. | Mar 2013 | B2 |
8934244 | Shelnutt et al. | Jan 2015 | B2 |
20040190247 | Chu et al. | Sep 2004 | A1 |
20050248922 | Chu et al. | Nov 2005 | A1 |
20080067805 | Kamada et al. | Mar 2008 | A1 |
20080232064 | Sato et al. | Sep 2008 | A1 |
20090086428 | Campbell et al. | Apr 2009 | A1 |
20090086432 | Campbell et al. | Apr 2009 | A1 |
20090154096 | Iyengar et al. | Jun 2009 | A1 |
20090262501 | Claassen et al. | Oct 2009 | A1 |
20100103614 | Campbell | Apr 2010 | A1 |
20100103618 | Campbell et al. | Apr 2010 | A1 |
20130106265 | Shelnutt | May 2013 | A1 |
20130112378 | Shelnutt et al. | May 2013 | A1 |
20130312839 | Shelnutt et al. | Nov 2013 | A1 |
20130312846 | Eriksen et al. | Nov 2013 | A1 |
20130312854 | Eriksen et al. | Nov 2013 | A1 |
20140202678 | Goth | Jul 2014 | A1 |
20140203550 | Utsch | Jul 2014 | A1 |
20140218859 | Shelnutt et al. | Aug 2014 | A1 |
20140321056 | Yoshikawa et al. | Oct 2014 | A1 |
20150109735 | Campbell et al. | Apr 2015 | A1 |
20160066480 | Eckberg et al. | Mar 2016 | A1 |
20160242319 | Edwards et al. | Aug 2016 | A1 |
20160366792 | Smith | Dec 2016 | A1 |
20170049009 | Steinke | Feb 2017 | A1 |
Entry |
---|
Brown, Robert D., Notice of Allowance, U.S. Appl. No. 15/017,582, The United States Patent and Trademark Office, dated Jun. 13, 2017. |
Dravininkas, Adam B., Non-Final Office Action, U.S. Appl. No. 15/016,249, The United States Patent and Trademark Office, dated Jun. 20, 2017. |
Number | Date | Country | |
---|---|---|---|
20170181323 A1 | Jun 2017 | US |
Number | Date | Country | |
---|---|---|---|
62270563 | Dec 2015 | US |