# US 2024/0004447 A1 · Power throttling of high performance computing (hpc) platform components

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0004447 A1
- Document ID: US20240004447A1
- Status: Published application
- Kind: A1
- Application number: 18/468,771
- Filed: 2023-09-18
- Published: 2024-01-04
- Earliest claimed: 2021-10-15
- Assignees: Dell Products L.P.
- Applicants: Dell Products, L.P.
- Inventors: Akkiah Choudary Maddukuri; Timothy M. Lambert; Elie Antoun Jreij; Bhavesh Govindbhai Patel; Mukund P. Khatri
- Web page: https://patentgrep.com/patent/US20240004447A1

## CPC

- G06F 1/206: {comprising thermal management}
- G06F 1/26: Power supply means, e.g. regulation thereof (for memories G11C)
- G06F 1/30: Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations (for resetting only G06F1/24)
- G06F 1/3206: Monitoring of events, devices or parameters that trigger a change in power modality
- G06F 1/324: by lowering clock frequency
- G06F 1/3287: by switching off individual functional units in the computer system
- G06F 1/3296: by lowering the supply or operating voltage
- G06F 9/4893: {taking into account power or heat criteria (power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206)}
- G06F 9/5094: {where the allocation takes into account power or heat criteria (power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206)}
- G06F 1/28: Supervision thereof, e.g. detecting power-supply failure by out of limits supervision
- G06F 11/3062: {where the monitored property is the power consumption (power management in a computing system G06F1/3203)}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

Embodiments of systems and methods for power throttling of High Performance Computing (HPC) components are described. In some embodiments, an HPC platform may include: a system Baseboard Management Controller (BMC), and an accelerator tray comprising a tray BMC coupled to a plurality of managed subsystems and to the system BMC, where the system BMC is configured to: in response to a power excursion event, instruct the tray BMC to throttle a first managed subsystem by a first amount and to throttle a second managed subsystem by a second amount.

## Claims

### Claim 1

A first Baseboard Management Controller (BMC) of an Information Handling System (IHS), the first BMC having program instructions stored thereon that, upon execution, cause the first BMC to:
determine a power consumption contribution of each of a plurality of managed subsystems coupled to a second BMC; and
in response to a power excursion event, throttle one or more of the plurality of managed subsystems, at least in part, based upon the power consumption contributions.

### Claim 2 (depends on 1)

The first BMC of claim 1, wherein the first BMC is coupled to a host processor of the HPC platform, wherein a second BMC is decoupled from the host processor, and wherein the first BMC and the second BMC are coupled to each other via at least one of: a Peripheral Component Interconnect Express (PCIe), RMII-Based Transport (RBT), I2C, or Universal Serial Bus (USB) link.

### Claim 3 (depends on 1)

The first BMC of claim 1, wherein the program instructions, upon execution, further cause the second BMC to throttle the one or more of the plurality of managed subsystems based upon at least one of: a priority of a workload assigned to each of the one or more of the plurality of managed subsystems, or a priority of each of the one or more of the plurality of managed subsystems.

### Claim 4

A method, comprising:
receiving, at a second Baseboard Management Controller (BMC) from a first BMC, a signal to throttle one or more of a plurality of managed subsystems coupled to the second BMC; and
throttling a first managed subsystem based upon at least one of: a policy or a license, wherein the first managed subsystem is throttled by an amount different than a second managed subsystem.

### Claim 5 (depends on 4)

The method of claim 4, wherein the signal comprises a PWRBRK signal, and wherein the method further comprises, in response to receiving the PWRBRK signal, performing a responsive action other than an emergency power reduction with respect to one or more of the plurality of managed subsystems.

### Claim 6

A High Performance Computing (HPC) platform, comprising:
a first Baseboard Management Controller (BMC); and
an accelerator tray comprising a second BMC coupled to a plurality of managed subsystems and to the first BMC, wherein the first BMC is configured to, in response to a power excursion event, instruct the second BMC to throttle a first managed subsystem by a first amount and to throttle a second managed subsystem by a second amount.

### Claim 7 (depends on 6)

The HPC platform of claim 6, wherein the first and second managed subsystems comprise at least one of: a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), an Intelligence Processing Unit (IPU), a Data Processing Unit (DPU), a Gaussian Neural Accelerator (GNA), an Audio and Contextual Engine (ACE), or a Vision Processing Unit (VPU).

### Claim 8 (depends on 6)

The HPC platform of claim 6, wherein the power excursion event is triggered by the detection of at least one of: a high output power supply current, a power supply input undervoltage, or an over temperature condition.

### Claim 9 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to identify an order in which each of the plurality of managed subsystems is to be throttled based, at least in part, upon a policy or a license.

### Claim 10 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to identify an order in which each of the plurality of managed subsystems is to be throttled based, at least in part, upon a priority of a workload.

### Claim 11 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to identify an order in which each of the plurality of managed subsystems is to be throttled based, at least in part, upon a priority of each managed subsystem.

### Claim 12 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to determine at least one of the first or second amounts based, at least in part, upon a policy or a license.

### Claim 13 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to determine the first amount based upon a first priority of a first workload assigned to the first managed subsystem and the second amount based upon a second priority of a second workload assigned to the second managed subsystem.

### Claim 14 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to determine the first amount based upon a first priority of the first managed subsystem and the second amount based upon a second priority of the second managed subsystem.

### Claim 15 (depends on 14)

The HPC platform of claim 14, wherein the first BMC is configured to communicate the first and second priorities to a workload manager.

### Claim 16 (depends on 15)

The HPC platform of claim 15, wherein the workload manager is configured to assign a high-priority workload to a last-to-throttle managed subsystem.

### Claim 17 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to transmit, to the second BMC, an indication of a programmable action to be performed with respect to the plurality of managed subsystems in response to a PWRBRK signal.

### Claim 18 (depends on 6)

The HPC platform of claim 6, further comprising a host processor coupled to the first BMC and decoupled from the second BMC, wherein the first BMC is configured to instruct the host processor to enter a throttling mode, at least in part, in response to the first and second amounts being insufficient to maintain a total power consumption of the HPC platform below a power cap.

### Claim 19 (depends on 6)

The HPC platform of claim 6, wherein the first BMC is configured to determine the power consumption contribution of each of the plurality of managed subsystems based upon power consumption information received from at least one of: (a) the second BMC via an Out-Of-Band (OOB) management link, or (b) directly from the plurality of managed subsystems.

### Claim 20 (depends on 19)

The HPC platform of claim 19, wherein the first BMC is configured to trigger migration of a workload executed by the first managed subsystem to the second managed subsystem based, at least in part, upon the power consumption information.
