IP Library Granted Patent US 12,248,352
Granted Patent B2
US 12,248,352 · App. 18/255,064 · Granted Mar 11, 2025

Computer orchestration

Inventors: Peter Willis (London, GB); Andrew Reid (London, GB)
Assignee: BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY
G06F1/324G06F1/3206G06F9/45558H04L41/40H04L67/10G06F1/3203G06F2009/45595
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,248,352
App. No.
18/255,064
Granted
Mar 11, 2025
Kind
B2
Abstract

A computer system comprising a plurality of computers, each computer comprising at least one processor, respectively, such that the computer system comprises a plurality of processors, and an orchestration system comprising an orchestrator. The orchestration system is configured to: forecast a traffic load of the plurality of processors to obtain a forecast traffic load: determine, in dependence on the forecast traffic load, a clock frequency for each of the plurality of processors so as to decrease a power consumption of the computer system; and using the orchestrator, instruct adjustment of the clock frequency for each respective processor of the plurality of processors to the clock frequency determined for the respective processor. This patent application further relates to a telecommunications network comprising a computer system, and an orchestration system.

Claims (47)

1. A computer system comprising:

a plurality of computers, each computer comprising at least one processor, respectively, such that the computer system comprises a plurality of processors; and

an orchestration system comprising an orchestrator, the orchestration system configured to:

forecast a traffic load of the plurality of processors to obtain a forecast traffic load;

determine, in dependence on the forecast traffic load, a clock frequency for each of the plurality of processors so as to decrease a power consumption of the computer system;

using the orchestrator, instruct adjustment of the clock frequency for each respective processor of the plurality of processors to the clock frequency determined for the respective processor;

receive an indication of a lack of access of at least one of the plurality of computers to a mains power source; and

in response to the indication, determine an updated clock frequency for at least one of the plurality of processors so as to increase a time over which each respective computer of the at least one of the plurality of computers is operational, using a power source comprised by the respective computer.

2. The computer system according to claim 1 , wherein the orchestration system is configured to determine the clock frequency for each of the plurality of processors in further dependence on a workflow to be performed by the plurality of processors.

3. The computer system according to claim 2 , wherein the workflow indicates that performance of a second task by a second processor of the plurality of processors depends on a performance of at least part of a first task by a first processor of the plurality of processors, wherein optionally the orchestration system is configured to determine to increase the clock frequency of the first processor in response to determining that the workflow indicates that the performance of the second task by the second processor depends on the performance of at least part of the first task by the first processor.

4. The computer system according to claim 2 , wherein the orchestrator is configured to generate the workflow.

5. The computer system according to claim 1 , wherein the orchestration system is configured to determine the clock frequency for each of the plurality of processors in further dependence on a topology of the computer system.

6. The computer system according to claim 1 , wherein:

the clock frequency for each of the plurality of processors comprises a second clock frequency for a processor of the plurality of processors;

the orchestration system comprises a frequency determination component for determining, in dependence on the forecast traffic load, a first clock frequency for the processor so as to decrease the power consumption of the computer system; and

the orchestrator is configured to:

determine the second clock frequency for the processor, in dependence on the first clock frequency and characteristic data indicative of at least one of: a characteristic of a service to be provided or a characteristic of the computer system; and

instruct the adjustment of the clock frequency for the processor to the second clock frequency.

7. The computer system according to claim 1 , wherein the orchestration system is configured to determine, in dependence on the forecast traffic load, that a clock frequency of a processor of the plurality of processors is to be increased so as to decrease the power consumption of the computer system.

8. The computer system according to claim 7 , wherein the orchestration system is configured to determine that the clock frequency of the processor is to be increased relative to a further processor of the plurality of processors such that a rate of performance of a task by the processor is greater than a rate of performance of a further task by the further processor.

9. The computer system according to claim 1 , wherein the orchestration system is configured to orchestrate the implementation of at least one virtual network function (VNF) using at least one of the plurality of processors.

10. The computer system according to claim 9 , wherein determining the clock frequency for each of the plurality of processors comprises determining a first clock frequency for a first processor of the plurality of processors to implement a first VNF of the at least one VNF and a second clock frequency, different from the first clock frequency, for a second processor of the plurality of processors to implement a second VNF of the at least one VNF.

11. The computer system according to claim 9 , wherein the orchestration system is configured to determine, in dependence on the forecast traffic load, a count of virtual machine (VM) instances to instantiate to implement the at least one VNF, using the at least one of the plurality of processors.

12. The computer system according to claim 9 , wherein:

the clock frequency for each of the plurality of processors comprises a, or the, second clock frequency for each of the at least one of the plurality of processors;

the orchestration system is configured to deploy, before the forecast traffic load is forecast to occur, at least one virtual machine (VM) for implementing the at least one VNF, using the at least one of the plurality of processors with respective first clock frequencies, and

the adjustment of the clock frequency comprises, when the forecast traffic load is forecast to occur, adjusting the clock frequency for each respective processor of the at least one of the plurality of processors to the second clock frequency determined for the respective processor, wherein, for each of the at least one of the plurality of processors, the second clock frequency is higher than the first clock frequency.

13. The computer system according to claim 1 , comprising a virtual infrastructure manager (VIM) to:

receive a command from the orchestrator to perform the adjustment of the clock frequency for each respective processor; and

send instructions to each of the plurality of computers, respectively, to instruct the adjustment of the clock frequency of the at least one processor of the respective computer.

14. The computer system according to claim 1 , wherein the orchestration system is configured to determine the clock frequency for each of the plurality of processors in further dependence on at least one of: a forecast energy usage of the plurality of processors or an environmental condition of an environment of the plurality of computers.

15. The computer system according to claim 14 , wherein the orchestration system is configured to forecast energy usage of the plurality of processors in dependence on a forecast temperature of the environment of the plurality of computers, to obtain the forecast energy usage.

16. The computer system according to claim 1 , wherein the orchestration system is configured to:

determine that a respective outcome of implementing at least one workload using the plurality of processors is independent of a clock frequency of the plurality of processors;

in response, perform the determining the clock frequency for each of the plurality of processors and the instructing the adjustment of the clock frequency for each respective processor of the plurality of processors; and

instruct the plurality of processors to implement the at least one workload.

17. The computer system according to claim 1 , wherein the orchestration system is configured to determine the clock frequency for each of the plurality of processors in further dependence on a forecast length of time for redistributing at least one existing workload of at least one of the plurality of processors, wherein optionally the orchestration system is configured to determine, in dependence on the forecast length of time for redistributing the at least one existing workload, that the clock frequency for the plurality of processors is to be adjusted instead of redistributing the at least one existing workload.

18. The computer system according to claim 1 , wherein the orchestration system is remote from the plurality of computers.

19. The computer system according to claim 1 , wherein the plurality of computers are distributed computers.

20. A telecommunications network comprising the computer system according to claim 1 , wherein at least one of the plurality of computers is configured to host a virtual machine for implementing a virtual network function (VNF).

21. A computer system comprising:

a plurality of computers, each computer comprising at least one processor, respectively, such that the computer system comprises a plurality of processors; and

an orchestration system comprising an orchestrator, the orchestration system configured to:

forecast a traffic load of the plurality of processors to obtain a forecast traffic load;

determine, in dependence on the forecast traffic load, a clock frequency for each of the plurality of processors so as to decrease a power consumption of the computer system; and

using the orchestrator, instruct adjustment of the clock frequency for each respective processor of the plurality of processors to the clock frequency determined for the respective processor;

wherein determining the clock frequency comprises determining the clock frequency in further dependence on a clock frequency adjustment policy, wherein optionally the clock frequency adjustment policy indicates a prioritisation level associated with: respective workloads to be implemented, respective tenants each associated with a respective workload to be implemented, and/or respective users each associated with a respective workload to be implemented.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 30, 2023
From: WILLIS, PETER; REID, ANDREW
To: BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY
Reel/Frame 063799/0524 →
Priority Claims (1)
GB 2018987 · Dec 2, 2020 · national
Continuity (1)
Related Publication 20240019923A1 · Jan 18, 2024
References Cited (26)
US 9575536B2 · Rajappa · 2017 [cited by examiner]
US 9696786B2 · Panda · 2017 [cited by examiner]
US 10289184B2 · Malik · 2019 [cited by examiner]
US 20160057075A1 · Parikh · 2016 [cited by examiner]
US 20170005515A1 · Sanders · 2017 [cited by examiner]
US 20170230267A1 · Armolavicius · 2017 [cited by examiner]
US 20180026906A1 · Balle et al. · 2018 [cited by applicant]
US 20180027060A1 · Metsch · 2018 [cited by examiner]
US 20180359029A1 · Shiner · 2018 [cited by examiner]
US 20190272002A1 · Seenappa · 2019 [cited by examiner]
US 20200106714A1 · Cote · 2020 [cited by examiner]
US 20200159281A1 · Seenappa et al. · 2020 [cited by applicant]
US 20200167258A1 · Chattopadhyay · 2020 [cited by examiner]
US 20200195528A1 · Barton · 2020 [cited by examiner]
US 20200310394A1 · Wouhaybi · 2020 [cited by examiner]
US 20210176189A1 · Rabipour · 2021 [cited by examiner]
US 20220066499A1 · Trim · 2022 [cited by examiner]
US 20220124009A1 · Metsch · 2022 [cited by examiner]
US 20240019923A1 · Willis · 2024 [cited by examiner]
US 20240276399A1 · Vasudevan · 2024 [cited by examiner]
CN 106569575A · 2017 [cited by applicant]
Combined Search and Examination Report for GB2018987.4, dated Aug. 27, 2021, 11 pages. [cited by applicant]
Ewa Niewiadomska-Szynkiewicz et al., “Resource Management System for HPC Computing”, Mar. 2018, ResearchGate, 11 pages, at URL: https://www.resarchgate.net/publication/323611448. [cited by applicant]
International Search Report and Written Opinion of the ISA for PCT/EP2021/081594 mailed Jan. 31, 2022, 18 pages. [cited by applicant]
Viresh Kumar, “CPU frequency governors and remote callbacks”, Sep. 4, 2017, LWN.net, 3 pages, at URL: https://lwn.net/Articles/732740/. [cited by applicant]
International Preliminary Report on Patentability dated Jun. 15, 2023, issued for International Application No. PCT/EP2021/081594 (10 pages). [cited by applicant]