WO2022271184A1

WO2022271184A1 - Redundant power supply unit connections

Info

Publication number: WO2022271184A1
Application number: PCT/US2021/039092
Authority: WO
Inventors: Chao-wen CHENG; Yu-Fang Chen; Tsung-Yen Chen; Wen-Bin Lin
Original assignee: Hewlett-Packard Development Company, L.P.
Priority date: 2021-06-25
Filing date: 2021-06-25
Publication date: 2022-12-29

Abstract

In an example in accordance with the present disclosure, a computing system is described. The computing system includes a first computing device. The first computing device includes a first power supply unit (PSU) to convert alternating current (AC) into direct current (DC) for use by components of the first computing device. The computing system also includes a redundant PSU. The computing system also includes a power distribution system to selectively couple the redundant PSU of the first computing device to a second computing device.

Description

REDUNDANT POWER SUPPLY UNIT CONNECTIONS

BACKGROUND

[0001] Computing devices are used by millions of people daily to carry out business, personal, and social operations. Examples of computing devices include desktop computers, laptop computers, all-in-one devices, servers, and gaming systems to name a few. While particular reference is made to a few types of computing devices, there are innumerable types of computing devices to which the current specification may apply.

BRIEF DESCRIPTION OF THE DRAWINGS

[0002] The accompanying drawings illustrate various examples of the principles described herein and are part of the specification. The illustrated examples are given merely for illustration, and do not limit the scope of the claims.

[0003] Fig. 1 is a block diagram of a computing system for sharing redundant power supply units (PSUs), according to an example of the principles described herein.

[0004] Fig. 2 is a diagram of a computing system for sharing redundant PSUs, according to an example of the principles described herein.

[0005] Fig. 3 is a flowchart of a method for sharing redundant PSUs, according to an example of the principles described herein.

[0006] Fig. 4 is a diagram of a computing system for sharing redundant PSUs, according to an example of the principles described herein. [0007] Fig. 5 is a diagram of a computing system for sharing redundant PSUs, according to an example of the principles described herein.

[0008] Fig. 6 is a flowchart of a method for sharing redundant PSUs, according to an example of the principles described herein.

[0009] Fig. 7 depicts a non-transitory machine-readable storage medium for sharing redundant PSUs, according to an example of the principles described herein.

[0010] Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements. The figures are not necessarily to scale, and the size of some parts may be exaggerated to more clearly illustrate the example shown. Moreover, the drawings provide examples and/or implementations consistent with the description; however, the description is not limited to the examples and/or implementations provided in the drawings.

DETAILED DESCRIPTION

[0011] Computing devices are used in many day-to-day personal and professional environments. For example, users may utilize computing devices in their own homes as well as in their professional lives. In fact, it is not uncommon for users to spend hours in front of a computer screen and to interact with multiple computing devices over the course of the day. However, it may be the case that in some environments, computing device developments may make them more suited to deliver an intended functionality.

[0012] For example, desktop computing devices, gaming systems, and server blades may include a power supply unit (PSU). The power supply unit (PSU) of a computing device converts high voltage alternating current (AC), which may be provided by an electric grid, into low-voltage direct current (DC). The PSU of a computing device may also regulate the DC output voltage to the tolerances used by computing devices and their associated hardware components. In some examples, such regulation may be done by controlling voltage, which might change automatically or manually depending on the PSU. That is, the PSU converts the power provided from the outlet into usable power for the many parts inside the computing device.

[0013] A PSU may include an electrical connection to the grid, a fan to cool the PSU, and switches to change the power supply voltage and turn the PSU on and off. Internally, the PSU may include a rectifier to convert AC into DC, a filter to smooth out the DC coming from a rectifier, a transformer to control the incoming voltage by stepping it up or down, and a voltage regulator to control the DC output, allowing the correct amount of power, volts, or watts, to be supplied to the computer hardware. The PSU also includes cables connected to internal components of the computing device including the motherboard.

As such, the components within the computing device chassis are powered by the PSU. For example, the motherboard, random access memory (RAM), central processing unit (CPU), hard drive, disk drives, and video cards among others may be drawing power from the PSU.

[0014] However, as with all computing device components, PSUs may fail. When a PSU fails, operations of the computing device cease. In some cases, PSU failure and the cessation of computing device operations may result in irretrievable loss of data and/or work product. Accordingly, the present specification describes the use of redundant PSUs in a computing device to provide greater protection against PSU failure. In this example, a computing device may include a main PSU and a redundant PSU. As such, if one PSU fails, the other can carry the load of providing power to components of the computing device.

[0015] However, to provide PSU redundancy, each computing device may have a dedicated main PSU as well as a redundant PSU. For example, two computing devices may each have two PSUs for a total of four PSUs to provide redundancy to both computing devices. Accordingly, the present specification describes a system to provide redundancy for both the computing devices, but reducing the quantity of PSUs coupled to the computing devices. For example, in a particular case, a first computing device may be a computer and a second computing device may be a peripheral component interconnect express (PCIe) expansion chassis that provides for high-speed connectivity between the computer and PCIe components. In this example, both the computer and the PCIe expansion chassis may have two PSUs for a total of four PSUs in a single computer enclosure. The present specification allows for both the computing device and the expansion chassis to have redundancy all while reducing the total count of PSUs in the computer enclosure. While specific reference is made to a computing device and second computing device environment, i.e., a personal computer and PCIe expansion slot, other environments may implement the computing system as described herein. For example, the second computing device may be a second personal computer or a server.

[0016] Specifically, the present specification describes a computing system that shares a redundant PSU between multiple computing devices. For example, a first computing device may have a main PSU and a redundant PSU while a second computing device may have a single PSU. In this example, the computing system includes a power distribution system to couple the redundant PSU from the first computing device to the second computing device. As such, a total of three PSUs (i.e., main PSU of first computing device, main PSU of second computing device, and redundant PSU of the first computing device) are used to provide redundancy to a two-computing device environment, whereas without such a system a total of four PSUs would be implemented to provide this level of redundancy. Still further, the present system that implements such a power distribution system may allow both, or either, of the computing devices to operate at higher levels, while still maintaining PSU redundancy.

[0017] Specifically, the present specification describes a computing system. The computing system includes a first computing device that includes a first power supply unit (PSU) to convert alternating current (AC) into direct current (DC) for use by components of the first computing device. The computing system also includes a redundant PSU. The computing system also includes a power distribution system to selectively couple the redundant PSU of the first computing device to a second computing device.

[0018] The present specification also describes a method. According to the method, a first computing device is detected as having a redundant PSU in addition to a main PSU. Responsive to a detected redundant PSU on the first computing device, a second computing device is coupled to the redundant PSU and power is provided from the redundant PSU to both the first computing device and the second computing device.

[0019] The present specification also describes a non-transitory machine- readable storage medium encoded with instructions executable by a processor. The machine-readable storage medium includes instructions, executable by the processor to detect 1) that a first computing device having a main PSU and a redundant PSU and also 2) a state of the redundant PSU. The instructions are also executable by the processor to cause the processor to detect a power connection between the first computing device and an expansion computing device. Responsive to the detection of a redundant PSU on the first computing device and a power connection between the first computing device and the expansion computing device and based on the state of the redundant PSU, the instructions are executable by the processor to cause the processor to selectively couple the redundant PSU to the expansion computing device and provide power from the redundant PSU to both the first computing device and the second computing device.

[0020] In summary, using such a system, method, and machine-readable storage medium may, for example, 1) provide PSU redundancy with a reduced quantity of PSU devices as compared to a per-computing device redundancy system; 2) share PSU loads based on computing system specific criteria; 3) provide for selective coupling of one computing device to the PSU of another device; and 4) provide for higher performance, i.e., greater power consuming, of one or both of the computing devices that share the redundant PSU. However, it is contemplated that the devices disclosed herein may address other matters and deficiencies in a number of technical areas, for example.

[0021] As used in the present specification and in the appended claims, the term “a number of” or similar language is meant to be understood broadly as any positive number including 1 to infinity.

[0022] Turning now to the figures, Fig. 1 is a block diagram of a computing system (100) for sharing redundant power supply units (PSUs), according to an example of the principles described herein. The computing system (100) may include a first computing device (102). The first computing device (102) may be of a variety of types including a desktop computer, all-in-one device, gaming system, or a server blade.

[0023] In this example, the first computing device (102) may include a main power supply unit (PSU) (104-1). As described above, a PSU (104) is to convert high-voltage AC, such as may be provided when the first computing device (102) is plugged into an electrical outlet, into lower-voltage DC, that is usable by the hardware components of the first computing device (102). That is, as described above, computing devices (102) include hardware components that draw power to execute an intended functionality of the first computing device (102). For example, the first computing device (102) may include a central processing unit (CPU), graphics processing unit (GPU), input/output devices to allow a user to interact with the computing device (102), and memory devices such as hard drives, disk drives, and any number of peripheral devices that enhance the functionality of the computing device (102). Each of these hardware components rely on power to operate, power which is managed by the PSU. In an example, the main PSU (104-1) is received in a housing of the first computing device (102). Specifically, the computing device may have slots to receive various components, once of which may be the main PSU (104-1). In this example, the slot may have a form factor to match the main PSU (104-1). [0024] In addition to the main PSU (104-1), the computing system (100) may include a redundant PSU (104-2). In some examples, the redundant PSU (104- 2), like the main PSU (104-1 ) is housed within a slot of the housing of the first computing device (102). In other examples, the redundant PSU (104-2) may be external to the first computing device (102) housing, for example in a second housing. The redundant PSU (104-2) provides a backup to the main PSU (104- 1 ) and may operate at the same time as the main PSU (104-1 ) or may be triggered for operation due to some event wherein the main PSU (104-1) is not operating as intended. For example, it may be that the main PSU (104-1) is subject to preventative maintenance or the main PSU (104-1) may be malfunctioning. In either case, the redundant PSU (104-2) may provide the functionality of the main PSU (104-1), i.e. , power supplication to hardware components of the first computing device (102) in the event the main PSU (104- 1) is unable to do so. As such, the redundant PSU (104-2) provides backup for the main PSU (104-1). The redundant PSU (104-2) may provide the same functionality of the main PSU (104-1 ) at the same time as the main PSU (104- 1). That is, the redundant PSU (104-2) may provide a load sharing function relative to the main PSU (104-1).

[0025] As described above, the redundant PSU (104-2) may also provide a backup for the PSU (104) of another computing device. For example, the first computing device (102) may be connected to another standalone computing device or a computing device that is an expansion to the first computing device (102). As described above, an expansion chassis may be a computing device that is coupled to the first computing device (102) to provide additional functionality. In this example, the second computing device may similarly have a main PSU. Still in this example, it may be desirable for this second computing device to have a redundant PSU. Rather than including a redundant PSU per computing device, the present computing system (100) may share the redundant PSU (104-2) between multiple computing devices.

[0026] Accordingly, the computing system (100) includes a power distribution system (106) to selectively couple the redundant PSU (104-2) of the first electronic computing device (102) to a second computing device. For example, the second computing device may be coupled to the redundant PSU (104-2) responsive to detection of the redundant PSU (104-2), based on a state of the redundant PSU (104-2), and/or based on a power draw request from either the first computing device (102) or the second computing device.

[0027] Fig. 2 is a diagram of a computing system (100) for sharing redundant PSUs (104-2), according to an example of the principles described herein. In Fig. 2 and others, a redundant PSU (104-2) is identified as an “rPSU” as opposed to a main PSU which is identified as a “PSU.” As described above, the first computing device (102-1) may include a main PSU (104-1) and a redundant PSU (104-2). These PSUs (104) provide power to the hardware components (212-1) found in the first computing device (102-1). [0028] In this example, the first computing device (102-1) may be coupled to a second computing device (102-2), which second computing device (102-2) may be an expansion computing device to provide additional functionality to the first computing device (102-1). As with the first computing device (102-1), the second computing device (102-2) may include a main PSU (104-3) to provide power to hardware components (212-2) of the second computing device (102- 2). In this case, rather than including a second redundant PSU, the second computing device (102-2) may share the redundant PSU (104-2) coupled to the first computing device (102-1).

[0029] To facilitate such PSU sharing, the computing system (100) includes a power distribution system (106) that selectively couples the PSUs (104) of the different computing devices (102). Specifically, the power distribution system (106) may selectively couple a main power path, indicated in solid line, between the various PSUs (104) and the various computing devices (102). For example, the power distribution system (106) may include a switch that, based on any number of triggers, closes the main power path between the first computing device PSUs (104-1 , 104-2) and the second computing device (102-2) and the second computing device PSU (104-3) and the first computing device (102-1). As such, each computing device (102) has a main PSU (104-1 , 104-3) from which it draws power, and each has a redundant, and shared, PSU (104-2) to rely on.

[0030] In addition to selectively coupling the main power path (in solid line) between the computing devices (102) and PSUs (104), the power distribution system (106) may further couple an auxiliary power path (in dashed lines) between the first computing device (102-1) and the second computing device (102-2). The auxiliary power path may be referred to as stand-by power and may provide power to the computing devices (102) when the computing devices (102) are in a standby mode (i.e. , when the computing device is off, but plugged into AC power). In this example, the auxiliary power path may provide power to a controller of the power distribution system (106) as well as to other hardware components of the power distribution system (106), some examples of which are depicted in Fig. 5 below. [0031] In an example, the power distribution system (106) is disposed within the first computing device (102-1). In other examples, such as that depicted in Fig. 2, the power distribution system (106) is a separate device from the first computing device (102-1) and the second computing device (102-2) and is electrically connected to both via a cable connection.

[0032] Fig. 3 is a flowchart of a method (300) for sharing redundant PSUs (Fig. 1 , 104-2), according to an example of the principles described herein. As described above, there may be any number of triggers that initiate the selective coupling of the redundant PSU (Fig. 1 , 104-2) to the second computing device (Fig. 2, 102-2). For example, the method (300) may detect (block 301 ) that a first computing device (Fig. 1 , 102) includes, or is coupled to, a redundant PSU (Fig. 1 , 104-2). That is, electrical connection between the redundant PSU (Fig.

1 , 104-2), may transmit an electrical signal to the power distribution system (Fig. 1 , 106) so indicating.

[0033] In other examples, other criteria such as a power request from either computing device (Fig. 1 , 102), a state of the computing devices (Fig. 1 , 102), and/or a state of the PSUs (Fig. 1 , 104) themselves may trigger the coupling as described below.

[0034] In either example, responsive to a detected redundant PSU (Fig. 1 ,

104-2) on the first computing device (Fig. 2, 102-1 ), the power distribution system (Fig. 1 , 106) may selectively couple (block 302) a second computing device (Fig. 2, 102-2) to the redundant PSU (Fig. 1 , 104-2). In an example, this may include closing a switch along the main power path between the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102- 2) responsive to a detected redundant PSU (Fig. 1 , 104-2).

[0035] Responsive to the selective coupling, the power distribution system (Fig. 1 , 106) may provide (block 303) power from the redundant PSU (Fig. 1 , 104-2) to both the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102-2). In addition to providing the power, the power distribution system (Fig. 1 , 106) may provide a notification of the established power path between the first computing device (Fig. 2, 102-1) and the second, or expansion, computing device (Fig. 2, 102-2). As such, according to the method (300) both computing devices (Fig. 2, 102) are provided with redundant PSU support, in addition to a main PSU (Fig. 2, 104-1 , 104-3) that each computing device (Fig. 1 , 102) may have.

[0036] Fig. 4 is a diagram of a computing system (100) for sharing redundant PSUs (104-2, 104-4), according to an example of the principles described herein. In the example depicted in Fig. 4, in addition to the first redundant PSU (104-2) coupled to the first computing device (102-1), the computing system (100) may include a second redundant PSU (104-4) disposed within the second computing device (102-2). That is, the second computing device (102-2) housing may include an expansion slot wherein an additional PSU (104) may be inserted into the second computing device (102-2). In this example, in addition to selectively coupling the first redundant PSU (104-2) to the second computing device (102-2), the power distribution system (106) may selectively couple the second redundant PSU (104-2) to the first computing device (102-1).

[0037] In so doing, both of the computing devices (102-1) have redundant PSU support. This environment may also provide for increased processing at either the first computing device (102-1) or the second computing device (102- 2). For example, without the power distribution system (106), were the first computing device (102-1) provided with a 650 watt (W) main PSU (104-1), hardware components (212-1) of the first computing device (102-1) may consume 650 W of power, while maintaining redundancy via the first redundant PSU (104-2). Such hardware components (212-1) may also consume 1300 W of power from the first main PSU (104-1) and the first redundant PSU (104-2).

In other words, in order to consume more power than what the PSU (104) is rated, a computing device (102) may sacrifice redundancy. However, the arrangement as depicted in Fig. 4 facilitates consuming more power than what the PSUs (104) are rated for, all while maintaining redundancy.

[0038] For example, provided that each of the PSUs (104-1 , 104-2, 104-3, 104-4) are rated at 650 W, three of the PSUs (104-1 , 104-3, 104-4) may provide a total of 1 ,950 W to be evenly distributed among the two computing devices (102-1 , 102-2). That is, each computing device (102) may consume 975 W of power, which is more than the 650 W rating of each PSU (104). However, redundancy is not sacrificed as the remaining PSU (104-2) may be held in reserve as a redundant PSU (104). As such, the present arrangement as depicted in Fig. 4 provides PSU sharing and power consumption at power levels commensurate with the rating of the PSUs (104), or provides for power consumption at levels greater than the rating of the PSUs (104), with redundancy, albeit at a lower redundancy level. In an example, responsive to the selective coupling of computing devices (102) each of which have a main PSU and a redundant PSU, the computing system (100) may provide a notification of an increased power budget to each of the first computing device (102-1) and the second, or expansion, computing device (102-2). That is, each computing device (102) may be notified via hardware components, that a maximum amount of available power has been increased. As described in this example, whether consuming less than, the same amount, or more power than the PSUs (104) are rated for, the various PSUs (104-1 , 104-2, 104-3, 104-4) may equally share the load of the first computing device (102-1) and the second computing device (102-2).

[0039] Fig. 5 is a diagram of a computing system (100) for sharing redundant PSUs (104), according to an example of the principles described herein. Specifically, Fig. 5 depicts various components of the power distribution system (106). As described above, the power distribution system (106) may selectively couple the computing devices (102). Accordingly, the power distribution system (106) may include a switch (512) disposed along a main power path. In this example, the power distribution system (106) also includes a controller (514) to selectively close the switch (512) to selectively couple the redundant PSU (104- 2) of the first computing device (102-1) to the second computing device (102-2). In the case that the second computing device (102-2) also includes a redundant PSU, this closing of the switch (512) also selectively couples the redundant PSU of the second computing device (102-2) to the first computing device. [0040] While the main power path may be selectively coupled, the auxiliary power path may be continuously closed. As described above, this auxiliary power path may provide the controller (514) with power to operate. In this arrangement, both the first computing device (102-1) and the second computing device (102-2) receive standby power from an of the connected PSUs (104). [0041] In an example, the controller (514) may also receive information by which the switch (512) is closed. For example, the controller (514) may collect such information as power budgets from either computing device (102), PSU (104) status, etc. As an example, the connected state of a redundant PSU (104-2) may trigger closing of the switch (512). That is, a motherboard may detect connection of various components thereto. The power distribution system (106) which may be external to the first computing device (102-1) or connected to the first computing device (102-1) may receive such a detection signal for the redundant PSU (104-2) from the first computing device (102-1). Responsive to receipt of such a signal, the controller (514) may close the switch (512) to provide the power path between the redundant PSU (104-2) of the first computing device (102-1) and the second computing device (102-2).

[0042] In an example, in the event that the power budget, or power draw request for either computing device (102) does not necessitate, the switch (512) may be opened. In this example, the controller (514) may close the switch (512) when the power budget and/or power draw request dictates. The controller (514), via a connection to the first computing device (102-1) and the second computing device (102-2) may receive signals indicative of the power draw request of either computing device (102).

[0043] In another example, if the redundant PSU (104-2) is not healthy, i.e., has a flagged state, the controller (514) may open the switch (512) so as to prevent reliance of the second computing device (102-2) on a PSU (104-2) that is not functioning as intended. Similarly, such a flag may be transmitted from the redundant PSU (104-2) to the controller (514) via the first computing device (102-1). While specific examples are provided of status information that may be communicated to the controller (514) via either of the computing devices (102), other status information may be similarly shared which may trigger the selective coupling of a main power path between the computing devices (102).

[0044] The controller (514) may include a processor, an application-specific integrated circuit (ASIC), a semiconductor-based microprocessor, a central processing unit (CPU), and a field-programmable gate array (FPGA), and/or other hardware device and memory.

[0045] The memory may include a computer-readable storage medium, which computer-readable storage medium may contain, or store computer- usable program code for use by or in connection with an instruction execution system, apparatus, or device. The memory may include many types of memory including volatile and non-volatile memory. For example, the memory may include Random Access Memory (RAM), Read Only Memory (ROM), optical memory disks, and magnetic disks, among others. The executable code may, when executed by the respective component, cause the component to implement the functionality described herein.

[0046] The controller (514) may also provide signals indicative of the state of the power paths between the computing devices (102). Such notification may be provided to a user. Accordingly, the power distribution system (106) may include a display panel (516) to provide status information, either of the PSUs (104_, computing devices (102), power distribution system (106), and/or power distribution characteristics.

[0047] In some examples, the controller (514) includes a remote connection device (518). Via this remote connection, a user or administrator may view the collected information on the computing system (100) and update settings or manipulate operation of the power distribution system (106) or other components.

[0048] Fig. 6 is a flowchart of a method (600) for sharing redundant PSUs (Fig. 1 , 104), according to an example of the principles described herein. According to the method (600), the computing system (Fig. 1 , 100) detects (block 601 ) that a first computing device (Fig. 2, 102-1 ) includes a redundant PSU (Fig. 1 , 104-2). This may be performed as described above in connection with Fig. 3. The method (600) may also include detecting (block 602) a requested power draw from the first computing device (Fig. 2, 102-1) and a requested power draw from the second computing device (Fig. 2, 102-2). As described above, such information may be a trigger for determining whether to selectively close the switch (Fig. 5, 512) to establish PSU redundancy. [0049] Responsive to a detected redundant PSU, the power distribution system (Fig. 1 , 106) may selectively couple (block 603) the second computing device (Fig. 2, 102-2) to the redundant PSU (Fig. 1 , 104-2) and provide (block 604) power from the redundant PSU (Fig. 1 , 104-2) to the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102-2). These operations may be performed as described above in connection with Fig. 3. [0050] In an example, in addition to providing power to the computing devices (Fig. 1 , 102) from each computing device’s (Fig. 1, 102) respective main PSU (Fig. 1 , 104-1 , 104-3) and the redundant PSU (Fig. 1 , 104-2), the method (600) may include equally distributing (block 605) the power draw from the first computing device (Fig. 2, 102-1) and the second computing device (102-2). For example, in the event that there are two redundant PSUS (Fig. 1 , 104-2, 104-4), the method (600) may include equally distributing (block 605) the power draw from the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102-2) across a main PSU (Fig. 1 , 104-1) of the first computing device (Fig. 2, 102-1), a main PSU (Fig. 2, 104-3) of the second computing device (Fig. 2, 102-2), and the redundant PSU (Fig. 1 , 104-2). Doing so may ensure equal load across all the PSUs (Fig. 1, 104). Such load sharing measures may enhance the operation and life of each PSU (Fig. 1 , 104).

[0051] As described above, in some examples, there may be a second redundant PSU (Fig. 2, 104-4), which may be on the second computing device (Fig. 2, 102-2). As such, the method (600) includes the detection (block 606) of the second redundant PSU (Fig. 2, 104-4) on the second computing device (Fig. 2, 102-2) and triggering (block 607) an aggregated mode for the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102-2). As described above, in such an aggregated mode both the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102-2) power budget is increased to utilize the full capacity of the main PSU (Fig. 1 , 104-1) of the first computing device (Fig. 2, 102-1), a main PSU (Fig. 2, 104-3) of the second computing device (Fig. 2, 102-2), and the redundant PSU (Fig. 1 , 104-2) of the first computing device (Fig. 2, 102-1). In this example the second redundant PSU (Fig. 2, 104-4) on the second computing device (Fig. 2, 102-2) may provide redundant power.

[0052] Fig. 7 depicts a non-transitory machine-readable storage medium (720) for sharing redundant PSUs (Fig. 1 , 104), according to an example of the principles described herein. To achieve its desired functionality, the power distribution system (Fig. 1 , 106) includes various hardware components. Specifically, the power distribution system (Fig. 1 , 106) includes a processor and a machine-readable storage medium (720). The machine-readable storage medium (720) is communicatively coupled to the processor. The machine- readable storage medium (720) includes a number of instructions (722, 724,

726, 728, 730) for performing a designated function. In some examples, the instructions may be machine code and/or script code.

[0053] The machine-readable storage medium (720) causes the processor to execute the designated function of the instructions (722, 724, 726, 728, 730). The machine-readable storage medium (720) can store data, programs, instructions, or any other machine-readable data that can be utilized to operate the power distribution system (Fig. 1 , 106). Machine-readable storage medium (720) can store machine readable instructions that the processor of the power distribution system (Fig. 1 , 106) can process, or execute. The machine-readable storage medium (720) can be an electronic, magnetic, optical, or other physical storage device that contains or stores executable instructions. Machine- readable storage medium (720) may be, for example, Random-Access Memory (RAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a storage device, an optical disc, etc. The machine-readable storage medium (720) may be a non-transitory machine-readable storage medium (720).

[0054] Referring to Fig. 7, detect PSU instructions (722), when executed by the processor, cause the processor to, detect that a first computing device (Fig. 1 , 102) having a main PSU (Fig. 1 , 104-1) also has a redundant PSU (Fig. 1 , 104-2). Detect PSU state instructions (724), when executed by the processor, cause the processor to, detect a state of the redundant PSU (Fig. 1 , 104-2). Detect power connection instructions (726), when executed by the processor, cause the processor to, detect a power connection between the first computing device (Fig. 2, 102-1) and a second, or expansion, computing device (Fig. 2, 102-2). Couple PSU instructions (728), when executed by the processor, cause the processor to, responsive to a detected redundant PSU (Fig. 1 , 104-2) on the first computing device (Fig. 2, 102-1) and a power connection between the first computing device (Fig. 2, 102-1) and the expansion computing device and based on a state of the redundant PSU (Fig. 1 , 104-2), selectively couple the redundant PSU (Fig. 1 , 104-2) to the expansion computing device. Provide power instructions (730), when executed by the processor, cause the processor to, provide power from the redundant PSU (Fig. 1 , 104-2) to both the first computing device (Fig. 2, 102-1) and the second computing device (Fig. 2, 102- 2).

[0055] In summary, using such a system, method, and machine-readable storage medium may, for example, 1) provide PSU redundancy with a reduced quantity of PSU devices as compared to a per-computing device redundancy system; 2) share PSU loads based on computing system specific criteria; 3) provide for selective coupling of one computing device to the PSU of another device; and 4) provide for higher performance, i.e., greater power consuming, of one or both of the computing devices that share the redundant PSU. However, it is contemplated that the devices disclosed herein may address other matters and deficiencies in a number of technical areas, for example.

Claims

CLAIMS What is claimed is:

1 . A computing system, comprising: a first computing device comprising: a first power supply unit (PSU) to convert alternating current (AC) into direct current (DC) for use by hardware components of the first computing device; and a redundant PSU; and a power distribution system to selectively couple the redundant PSU of the first computing device to a second computing device.

2. The computing system of claim 1 , wherein the second computing device is an expansion computing device to provide additional functionality to the first computing device.

3. The computing system of claim 1 : further comprising a second redundant PSU disposed within the second computing device; and wherein the power distribution system is to selectively couple the second redundant PSU to the first computing device.

4. The computing system of claim 1 , wherein the power distribution system comprises: a switch disposed along a main power path between the first computing device and the second computing device; and a controller to selectively close the switch to selectively couple the redundant PSU of the first computing device to the second computing device.

5. The computing system of claim 4, wherein: the power distribution system further comprises an auxiliary power path between the first computing device and the second computing device; and the auxiliary power path is to provide power to the controller.

6. The computing system of claim 4, wherein the power distribution system further provides a display panel to provide status information regarding the power distribution system.

7. The computing system of claim 4, wherein the power distribution system further comprises a remote connection device.

8. A method, comprising: detecting that a first computing device comprises a redundant power supply unit (PSU), which redundant PSU is in addition to a main PSU; responsive to a detected redundant PSU on the first computing device, selectively coupling a second computing device to the redundant PSU; and providing power from the redundant PSU to both the first computing device and the second computing device.

9. The method of claim 8, further comprising detecting a requested power draw from the first computing device and a requested power draw from the second computing device.

10. The method of claim 9, wherein selectively coupling of the redundant power supply to the second computing device is based on requested power draws.

11. The method of claim 8, further comprising equally distributing the power draw from the first computing device and the second computing device across a main PSU of the first computing device, a main PSU of the second computing device, and the redundant PSU.

12. The method of claim 8, further comprising: detecting a second redundant PSU on the second computing device; and triggering an aggregated mode for the first computing device and the second computing device, wherein: both the first computing device and the second computing device power budget is increased to utilize a full capacity of the main PSU of the first computing device, a main PSU of the second computing device, and the redundant PSU of the first computing device; and the second redundant PSU provides redundant power supply.

13. A non-transitory machine-readable storage medium encoded with instructions executable by a processor, the machine-readable storage medium comprising instructions to: detect that a first computing device comprises a main power supply unit (PSU) and a redundant PSU; detect a state of the redundant PSU; detect a power connection between the first computing device and an expansion computing device; responsive to a detected redundant PSU on the first computing device and a power connection between the first computing device and the expansion computing device and based on the state of the redundant PSU, selectively couple the redundant PSU to the expansion computing device; and provide power from the redundant PSU to both the first computing device and the expansion computing device.

14. The non-transitory machine-readable storage medium of claim 13, further comprising instructions executable by a processor to, responsive to a selective coupling of the redundant PSU to the expansion computing device, provide a notification of an increased power budget for the first computing device and the expansion computing device.

15. The non-transitory machine-readable storage medium of claim 13, further comprising instructions executable by a processor to provide a notification of a power path between the first computing device and the expansion computing device.