JP7379668B2 - 機械学習ワークロードのためのタスクスケジューリング - Google Patents

機械学習ワークロードのためのタスクスケジューリング Download PDF

Info

Publication number
JP7379668B2
JP7379668B2 JP2022514245A JP2022514245A JP7379668B2 JP 7379668 B2 JP7379668 B2 JP 7379668B2 JP 2022514245 A JP2022514245 A JP 2022514245A JP 2022514245 A JP2022514245 A JP 2022514245A JP 7379668 B2 JP7379668 B2 JP 7379668B2
Authority
JP
Japan
Prior art keywords
host
hosts
resources
hardware
task
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
JP2022514245A
Other languages
English (en)
Japanese (ja)
Other versions
JP2023511467A (ja
Inventor
ワン，ジュエ
フアン，ホイ
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Google LLC
Original Assignee
Google LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Google LLC filed Critical Google LLC
Publication of JP2023511467A publication Critical patent/JP2023511467A/ja
Priority to JP2023187576A priority Critical patent/JP7637747B2/ja
Application granted granted Critical
Publication of JP7379668B2 publication Critical patent/JP7379668B2/ja
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5044—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering hardware capabilities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/502—Proximity
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/25—Using a specific main memory architecture
    • G06F2212/254—Distributed memory
    • G06F2212/2542—Non-uniform memory access [NUMA] architecture
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/20—Processor architectures; Processor configuration, e.g. pipelining

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Neurology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Advance Control (AREA)
  • Multi Processors (AREA)
JP2022514245A 2019-11-20 2020-09-08 機械学習ワークロードのためのタスクスケジューリング Active JP7379668B2 (ja)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2023187576A JP7637747B2 (ja) 2019-11-20 2023-11-01 機械学習ワークロードのためのタスクスケジューリング

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US201962938304P 2019-11-20 2019-11-20
US62/938,304 2019-11-20
US16/720,717 US11544113B2 (en) 2019-11-20 2019-12-19 Task scheduling for machine-learning workloads
US16/720,717 2019-12-19
PCT/US2020/049648 WO2021101617A1 (en) 2019-11-20 2020-09-08 Task scheduling for machine-learning workloads

Related Child Applications (1)

Application Number Title Priority Date Filing Date
JP2023187576A Division JP7637747B2 (ja) 2019-11-20 2023-11-01 機械学習ワークロードのためのタスクスケジューリング

Publications (2)

Publication Number Publication Date
JP2023511467A JP2023511467A (ja) 2023-03-20
JP7379668B2 true JP7379668B2 (ja) 2023-11-14

Family

ID=75910002

Family Applications (2)

Application Number Title Priority Date Filing Date
JP2022514245A Active JP7379668B2 (ja) 2019-11-20 2020-09-08 機械学習ワークロードのためのタスクスケジューリング
JP2023187576A Active JP7637747B2 (ja) 2019-11-20 2023-11-01 機械学習ワークロードのためのタスクスケジューリング

Family Applications After (1)

Application Number Title Priority Date Filing Date
JP2023187576A Active JP7637747B2 (ja) 2019-11-20 2023-11-01 機械学習ワークロードのためのタスクスケジューリング

Country Status (6)

Country Link
US (2) US11544113B2 (enExample)
EP (1) EP4062281A1 (enExample)
JP (2) JP7379668B2 (enExample)
KR (1) KR102759334B1 (enExample)
CN (1) CN114503077A (enExample)
WO (1) WO2021101617A1 (enExample)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11544113B2 (en) * 2019-11-20 2023-01-03 Google Llc Task scheduling for machine-learning workloads
WO2021195104A1 (en) * 2020-03-23 2021-09-30 Mentium Technologies Inc. Digital-imc hybrid system architecture for neural network acceleration
CN111930498B (zh) * 2020-06-29 2022-11-29 苏州浪潮智能科技有限公司 一种高效的gpu资源分配优化方法和系统
KR102871422B1 (ko) * 2020-10-21 2025-10-15 삼성전자주식회사 데이터 처리 방법 및 장치 및 이를 포함한 전자 장치 및 가속기 시스템
US11847489B2 (en) * 2021-01-26 2023-12-19 Apple Inc. United states graphics processor techniques with split between workload distribution control data on shared control bus and corresponding graphics data on memory interfaces
US11436054B1 (en) * 2021-04-05 2022-09-06 Hewlett Packard Enterprise Development Lp Directing queries to nodes of a cluster of a container orchestration platform distributed across a host system and a hardware accelerator of the host system
US11716257B1 (en) * 2021-05-24 2023-08-01 Neureality Ltd. Batching of artificial intelligence jobs
DE102021213282A1 (de) * 2021-11-25 2023-05-25 Robert Bosch Gesellschaft mit beschränkter Haftung Partizipatives Sicherheitsprotokoll für Datenwolken-basierte Funktionen
CN114490002B (zh) * 2022-02-17 2025-11-28 上海阵量智能科技有限公司 数据处理系统、任务调度方法、装置、芯片、及电子设备
US12417047B2 (en) * 2023-01-10 2025-09-16 Google Llc Heterogeneous ML accelerator cluster with flexible system resource balance
US20250045099A1 (en) * 2023-08-02 2025-02-06 Samsung Electronics Co., Ltd. Systems, methods, and apparatus for assigning machine learning tasks to compute devices
US12481534B2 (en) * 2024-01-22 2025-11-25 Dropbox, Inc. Dynamically selecting artificial intelligence models and hardware environments to execute tasks
TWI897553B (zh) * 2024-06-19 2025-09-11 聯發科技股份有限公司 處理單元的排程方法及非暫態機器可讀介質

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015132887A (ja) 2014-01-09 2015-07-23 富士通株式会社 要求分散プログラム、要求分散方法および情報処理装置
US20190114534A1 (en) 2017-10-17 2019-04-18 Xilinx, Inc. Neural network processing system having multiple processors and a neural network accelerator
US20190312772A1 (en) 2018-04-04 2019-10-10 EMC IP Holding Company LLC Topology-aware provisioning of hardware accelerator resources in a distributed environment
US20190325302A1 (en) 2018-04-23 2019-10-24 EMC IP Holding Company LLC Implementing parameter server in networking infrastructure for high-performance computing

Family Cites Families (37)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1476834A1 (en) 2002-02-07 2004-11-17 Thinkdynamics Inc. Method and system for managing resources in a data center
JP5056345B2 (ja) 2007-10-29 2012-10-24 富士通株式会社 データ処理装置およびデータ処理方法
US9652372B2 (en) 2010-12-15 2017-05-16 At&T Intellectual Property I, L.P. Method and apparatus for improving non-uniform memory access
US20130185729A1 (en) 2012-01-13 2013-07-18 Rutgers, The State University Of New Jersey Accelerating resource allocation in virtualized environments using workload classes and/or workload signatures
FR2991074B1 (fr) 2012-05-25 2014-06-06 Bull Sas Procede, dispositif et programme d'ordinateur de controle dynamique de distances d'acces memoire dans un systeme de type numa
JP5949188B2 (ja) 2012-06-08 2016-07-06 日本電気株式会社 密結合マルチプロセッサシステム
US9588804B2 (en) * 2014-01-21 2017-03-07 Qualcomm Incorporated System and method for synchronous task dispatch in a portable device
US9697045B2 (en) 2015-03-24 2017-07-04 International Business Machines Corporation Selecting resource allocation policies and resolving resource conflicts
US10241674B2 (en) 2015-12-11 2019-03-26 Vmware, Inc. Workload aware NUMA scheduling
US11153223B2 (en) * 2016-04-07 2021-10-19 International Business Machines Corporation Specifying a disaggregated compute system
WO2018067680A1 (en) 2016-10-05 2018-04-12 Hidden Path Entertainment, Inc. System and method of capturing and rendering a stereoscopic panorama using a depth buffer
US10176550B1 (en) 2017-03-20 2019-01-08 Nutanix, Inc. GPU resource usage display and dynamic GPU resource allocation in a networked virtualization system
CN107168782A (zh) * 2017-04-24 2017-09-15 复旦大学 一种基于Spark与GPU的并行计算系统
US11010205B2 (en) 2017-05-30 2021-05-18 Hewlett Packard Enterprise Development Lp Virtual network function resource allocation
US10686728B2 (en) * 2017-07-06 2020-06-16 Huawei Technologies Co., Ltd. Systems and methods for allocating computing resources in distributed computing
US10445249B2 (en) 2017-11-09 2019-10-15 International Business Machines Corporation Facilitating access to memory locality domain information
US10713092B2 (en) 2018-01-02 2020-07-14 Jpmorgan Chase Bank, N.A. Dynamic resource management of a pool of resources for multi-tenant applications based on sample exceution, query type or jobs
US10942767B2 (en) 2018-02-27 2021-03-09 Microsoft Technology Licensing, Llc Deep neural network workload scheduling
US10601903B2 (en) * 2018-05-17 2020-03-24 International Business Machines Corporation Optimizing dynamical resource allocations based on locality of resources in disaggregated data centers
US11030012B2 (en) 2018-09-28 2021-06-08 Intel Corporation Methods and apparatus for allocating a workload to an accelerator using machine learning
US11216314B2 (en) * 2018-11-02 2022-01-04 EMC IP Holding Company LLC Dynamic reallocation of resources in accelerator-as-a-service computing environment
US12254526B2 (en) * 2019-03-15 2025-03-18 Intel Corporation On chip dense memory for temporal buffering
US11184236B2 (en) * 2019-04-30 2021-11-23 Intel Corporation Methods and apparatus to control processing of telemetry data at an edge platform
US11521042B2 (en) 2019-05-21 2022-12-06 Anil Ravindranath System and method to dynamically and automatically sharing resources of coprocessor AI accelerators
US11301307B2 (en) * 2019-07-24 2022-04-12 Red Hat, Inc. Predictive analysis for migration schedulers
US12052260B2 (en) * 2019-09-30 2024-07-30 International Business Machines Corporation Scalable and dynamic transfer learning mechanism
US20210149677A1 (en) * 2019-11-15 2021-05-20 Intel Corporation Enhanced processor functions for calculation
US11726793B2 (en) * 2019-11-15 2023-08-15 Intel Corporation Data locality enhancement for graphics processing units
US11544113B2 (en) * 2019-11-20 2023-01-03 Google Llc Task scheduling for machine-learning workloads
US11586932B2 (en) * 2020-03-10 2023-02-21 International Business Machines Corporation Model training with variable batch sizing and gradient checkpoint segments
US11526964B2 (en) * 2020-06-10 2022-12-13 Intel Corporation Deep learning based selection of samples for adaptive supersampling
US12216738B2 (en) * 2020-10-12 2025-02-04 International Business Machines Corporation Predicting performance of machine learning models
WO2022118322A1 (en) * 2020-12-02 2022-06-09 Unifabrix Ltd. System and method for multimodal computer address space provisioning
US20220188691A1 (en) * 2020-12-11 2022-06-16 International Business Machines Corporation Machine Learning Pipeline Generation
US20220124005A1 (en) * 2021-11-16 2022-04-21 Kshitij Arun Doshi Systems and methods for reactive intent-driven end-to-end orchestration
US12461781B2 (en) * 2021-12-22 2025-11-04 Intel Corporation Low power inference engine pipeline in a graphics processing unit
US20230137191A1 (en) * 2022-11-12 2023-05-04 Adrian C. Hoban Mechanism to recompose workload packages in a computing environment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015132887A (ja) 2014-01-09 2015-07-23 富士通株式会社 要求分散プログラム、要求分散方法および情報処理装置
US20190114534A1 (en) 2017-10-17 2019-04-18 Xilinx, Inc. Neural network processing system having multiple processors and a neural network accelerator
US20190312772A1 (en) 2018-04-04 2019-10-10 EMC IP Holding Company LLC Topology-aware provisioning of hardware accelerator resources in a distributed environment
US20190325302A1 (en) 2018-04-23 2019-10-24 EMC IP Holding Company LLC Implementing parameter server in networking infrastructure for high-performance computing

Also Published As

Publication number Publication date
KR102759334B1 (ko) 2025-01-22
JP2024020271A (ja) 2024-02-14
JP7637747B2 (ja) 2025-02-28
WO2021101617A1 (en) 2021-05-27
US11544113B2 (en) 2023-01-03
KR20220038497A (ko) 2022-03-28
US20210149729A1 (en) 2021-05-20
JP2023511467A (ja) 2023-03-20
CN114503077A (zh) 2022-05-13
US12321781B2 (en) 2025-06-03
US20230136661A1 (en) 2023-05-04
EP4062281A1 (en) 2022-09-28

Similar Documents

Publication Publication Date Title
JP7379668B2 (ja) 機械学習ワークロードのためのタスクスケジューリング
CN111247533B (zh) 用于神经网络加速的机器学习运行时库
KR102860886B1 (ko) 스케줄러, 스케줄러의 동작 방법 및 이를 포함한 가속기 시스템
CN115269717A (zh) 存储设备、分布式存储系统以及数据处理方法
CN114730275A (zh) 使用张量在分布式计算系统中进行矢量化资源调度的方法和装置
US12314851B2 (en) Microservice-based training systems in heterogeneous graphic processor unit (GPU) cluster and operating method thereof
KR102871422B1 (ko) 데이터 처리 방법 및 장치 및 이를 포함한 전자 장치 및 가속기 시스템
US11954518B2 (en) User-defined metered priority queues
CN102929725A (zh) 信号处理并行计算软件的动态重配置方法
KR102742714B1 (ko) 중요 경로 기반 스케줄링 및 후처리를 통한 리소스 활용을 극대화할 수 있는 multi-GPU 기반 딥러닝 모델 추론 기법
US20220229695A1 (en) System and method for scheduling in a computing system
KR20230064963A (ko) 클러스터 컴퓨팅 시스템에서의 리소스 할당 방법 및 장치
KR20220049294A (ko) 스케줄러, 스케줄러의 동작 방법 및 이를 포함한 전자 장치
CN113841132B (zh) 用于主机设备接口的非对称数据通信
KR102787376B1 (ko) 가속기를 분할하는 전자 장치, 배치를 할당하는 전자 장치 및 그 동작 방법
CN116670649B (zh) 用于设计具有优化的系统延时的机器人系统架构的方法和系统
CN116166396A (zh) 调度模型的训练方法、装置、电子设备及可读存储介质
US11663465B2 (en) Method of managing task performance in an artificial neural network, and system executing an artificial neural network
WO2021199396A1 (ja) 分散処理ノードおよび分散処理システム
KR102831990B1 (ko) 이기종 그래픽 프로세서 유닛 클러스터를 이용한 마이크로서비스 기반의 트레이닝 시스템 및 그 동작 방법
HK40071346A (en) Task scheduling for machine-learning workloads
US12153957B2 (en) Hierarchical work scheduling
CN120407195B (zh) 混合异构计算任务的执行方法、设备、介质和程序产品
CN114461331B (zh) 一种资源预部署方法、装置、电子设备及可读存储介质
US20240095083A1 (en) Parallel workload scheduling based on workload data coherence

Legal Events

Date Code Title Description
A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20221121

A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20221121

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20230824

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20231003

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20231101

R150 Certificate of patent or registration of utility model

Ref document number: 7379668

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150