CA3021416C - Syncronization in a multi-tile processing arrangement - Google Patents

Syncronization in a multi-tile processing arrangement Download PDF

Info

Publication number
CA3021416C
CA3021416C CA3021416A CA3021416A CA3021416C CA 3021416 C CA3021416 C CA 3021416C CA 3021416 A CA3021416 A CA 3021416A CA 3021416 A CA3021416 A CA 3021416A CA 3021416 C CA3021416 C CA 3021416C
Authority
CA
Canada
Prior art keywords
tiles
instruction
tile
group
sync
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
CA3021416A
Other languages
English (en)
French (fr)
Other versions
CA3021416A1 (en
Inventor
Simon Christian Knowles
Alan Graham Alexander
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.)
Graphcore Ltd
Original Assignee
Graphcore Ltd
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 Graphcore Ltd filed Critical Graphcore Ltd
Publication of CA3021416A1 publication Critical patent/CA3021416A1/en
Application granted granted Critical
Publication of CA3021416C publication Critical patent/CA3021416C/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
    • G06F9/522—Barrier synchronisation
    • 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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30076—Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
    • G06F9/30087—Synchronisation or serialisation instructions
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163—Interprocessor communication
    • G06F15/173—Interprocessor communication using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
    • G06F15/17306—Intercommunication techniques
    • G06F15/17325—Synchronisation; Hardware support therefor
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/76—Architectures of general purpose stored program computers
    • G06F15/80—Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors
    • 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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30076—Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
    • G06F9/3009—Thread control instructions
    • 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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
    • G06F9/3851—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution from multiple instruction streams, e.g. multistreaming
    • 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
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3885—Concurrent instruction execution, e.g. pipeline or look ahead using a plurality of independent parallel functional units
    • 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
    • G06F9/461—Saving or restoring of program or task context
    • G06F9/462—Saving or restoring of program or task context with multiple register sets
    • 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
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • 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
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G06F9/4887—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues involving deadlines, e.g. rate based, periodic
    • 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
    • G06N3/10—Interfaces, programming languages or software development kits, e.g. for simulating neural networks
    • G06N3/105—Shells for specifying net layout
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • 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
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Advance Control (AREA)
  • Executing Machine-Instructions (AREA)
  • Multi Processors (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
CA3021416A 2017-10-20 2018-10-19 Syncronization in a multi-tile processing arrangement Active CA3021416C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GB1717291.7A GB2569269B (en) 2017-10-20 2017-10-20 Synchronization in a multi-tile processing arrangement
GB1717291.7 2017-10-20

Publications (2)

Publication Number Publication Date
CA3021416A1 CA3021416A1 (en) 2019-04-20
CA3021416C true CA3021416C (en) 2021-03-30

Family

ID=60481655

Family Applications (1)

Application Number Title Priority Date Filing Date
CA3021416A Active CA3021416C (en) 2017-10-20 2018-10-19 Syncronization in a multi-tile processing arrangement

Country Status (10)

Country Link
US (2) US10564970B2 (https=)
JP (1) JP6797881B2 (https=)
KR (1) KR102262483B1 (https=)
CN (1) CN110214317B (https=)
CA (1) CA3021416C (https=)
DE (1) DE102018126004A1 (https=)
FR (1) FR3072800B1 (https=)
GB (1) GB2569269B (https=)
TW (1) TWI700634B (https=)
WO (1) WO2019076714A1 (https=)

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2569098B (en) * 2017-10-20 2020-01-08 Graphcore Ltd Combining states of multiple threads in a multi-threaded processor
US11995448B1 (en) 2018-02-08 2024-05-28 Marvell Asia Pte Ltd Method and apparatus for performing machine learning operations in parallel on machine learning hardware
US12112175B1 (en) 2018-02-08 2024-10-08 Marvell Asia Pte Ltd Method and apparatus for performing machine learning operations in parallel on machine learning hardware
US10970080B2 (en) 2018-02-08 2021-04-06 Marvell Asia Pte, Ltd. Systems and methods for programmable hardware architecture for machine learning
DE102018205392A1 (de) * 2018-04-10 2019-10-10 Robert Bosch Gmbh Verfahren und Vorrichtung zur Fehlerbehandlung in einer Kommunikation zwischen verteilten Software Komponenten
DE102018205390A1 (de) * 2018-04-10 2019-10-10 Robert Bosch Gmbh Verfahren und Vorrichtung zur Fehlerbehandlung in einer Kommunikation zwischen verteilten Software Komponenten
US10929778B1 (en) 2018-05-22 2021-02-23 Marvell Asia Pte, Ltd. Address interleaving for machine learning
US11016801B1 (en) 2018-05-22 2021-05-25 Marvell Asia Pte, Ltd. Architecture to support color scheme-based synchronization for machine learning
US10997510B1 (en) 2018-05-22 2021-05-04 Marvell Asia Pte, Ltd. Architecture to support tanh and sigmoid operations for inference acceleration in machine learning
US10929779B1 (en) * 2018-05-22 2021-02-23 Marvell Asia Pte, Ltd. Architecture to support synchronization between core and inference engine for machine learning
GB2575294B8 (en) * 2018-07-04 2022-07-20 Graphcore Ltd Host Proxy On Gateway
GB2580165B (en) 2018-12-21 2021-02-24 Graphcore Ltd Data exchange in a computer with predetermined delay
KR102670905B1 (ko) * 2019-08-22 2024-05-31 구글 엘엘씨 전파 지연 감소
CN112416053B (zh) * 2019-08-23 2023-11-17 北京希姆计算科技有限公司 多核架构的同步信号产生电路、芯片和同步方法及装置
GB2590658A (en) * 2019-12-23 2021-07-07 Graphcore Ltd Communication in a computer having multiple processors
GB2590661B (en) * 2019-12-23 2022-02-09 Graphcore Ltd Sync network
GB2591106B (en) * 2020-01-15 2022-02-23 Graphcore Ltd Control of data transfer between processors
GB2593756B (en) * 2020-04-02 2022-03-30 Graphcore Ltd Control of data transfer between processing nodes
GB2596872B (en) * 2020-07-10 2022-12-14 Graphcore Ltd Handling injected instructions in a processor
GB2597078B (en) 2020-07-14 2022-07-13 Graphcore Ltd Communication between host and accelerator over network
US12495042B2 (en) * 2021-08-16 2025-12-09 Capital One Services, Llc Systems and methods for resetting an authentication counter

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH02114362A (ja) * 1988-10-24 1990-04-26 Nec Corp 並列演算装置
JP2001175618A (ja) * 1999-12-17 2001-06-29 Nec Eng Ltd 並列計算機システム
US8035650B2 (en) * 2006-07-25 2011-10-11 Qualcomm Incorporated Tiled cache for multiple software programs
US8046727B2 (en) * 2007-09-12 2011-10-25 Neal Solomon IP cores in reconfigurable three dimensional integrated circuits
US8866827B2 (en) 2008-06-26 2014-10-21 Microsoft Corporation Bulk-synchronous graphics processing unit programming
US8539204B2 (en) 2009-09-25 2013-09-17 Nvidia Corporation Cooperative thread array reduction and scan operations
US9081501B2 (en) * 2010-01-08 2015-07-14 International Business Machines Corporation Multi-petascale highly efficient parallel supercomputer
GB201001621D0 (en) * 2010-02-01 2010-03-17 Univ Catholique Louvain A tile-based processor architecture model for high efficiency embedded homogenous multicore platforms
US8713290B2 (en) 2010-09-20 2014-04-29 International Business Machines Corporation Scaleable status tracking of multiple assist hardware threads
US20120179896A1 (en) * 2011-01-10 2012-07-12 International Business Machines Corporation Method and apparatus for a hierarchical synchronization barrier in a multi-node system
CN104081781B (zh) 2012-01-20 2018-11-02 Ge视频压缩有限责任公司 一种编码器、解码器、编解码方法及用于传送解多工的方法
US10067768B2 (en) 2014-07-18 2018-09-04 Nvidia Corporation Execution of divergent threads using a convergence barrier
US9348658B1 (en) * 2014-12-12 2016-05-24 Intel Corporation Technologies for efficient synchronization barriers with work stealing support
US10101786B2 (en) * 2014-12-22 2018-10-16 Intel Corporation Holistic global performance and power management
US9747108B2 (en) 2015-03-27 2017-08-29 Intel Corporation User-level fork and join processors, methods, systems, and instructions
US10310861B2 (en) * 2017-04-01 2019-06-04 Intel Corporation Mechanism for scheduling threads on a multiprocessor
US10672175B2 (en) * 2017-04-17 2020-06-02 Intel Corporation Order independent asynchronous compute and streaming for graphics

Also Published As

Publication number Publication date
CA3021416A1 (en) 2019-04-20
KR20190044570A (ko) 2019-04-30
TW201923556A (zh) 2019-06-16
JP2019079528A (ja) 2019-05-23
JP6797881B2 (ja) 2020-12-09
DE102018126004A1 (de) 2019-04-25
CN110214317A (zh) 2019-09-06
US20200089499A1 (en) 2020-03-19
CN110214317B (zh) 2023-05-02
FR3072800B1 (fr) 2024-01-05
GB2569269B (en) 2020-07-15
GB201717291D0 (en) 2017-12-06
US11593185B2 (en) 2023-02-28
GB2569269A (en) 2019-06-19
TWI700634B (zh) 2020-08-01
KR102262483B1 (ko) 2021-06-08
WO2019076714A1 (en) 2019-04-25
US10564970B2 (en) 2020-02-18
US20190121641A1 (en) 2019-04-25
FR3072800A1 (fr) 2019-04-26

Similar Documents

Publication Publication Date Title
US11593185B2 (en) Synchronization in a multi-tile processing arrangement
US11113060B2 (en) Combining states of multiple threads in a multi threaded processor
US11586483B2 (en) Synchronization amongst processor tiles
US11023413B2 (en) Synchronization in a multi-tile, multi-chip processing arrangement
US10628377B2 (en) Synchronization in a multi-tile processing arrangement

Legal Events

Date Code Title Description
MPN Maintenance fee for patent paid

Free format text: FEE DESCRIPTION TEXT: MF (PATENT, 6TH ANNIV.) - STANDARD

Year of fee payment: 6

U00 Fee paid

Free format text: ST27 STATUS EVENT CODE: A-4-4-U10-U00-U101 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE REQUEST RECEIVED

Effective date: 20241007

U11 Full renewal or maintenance fee paid

Free format text: ST27 STATUS EVENT CODE: A-4-4-U10-U11-U102 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE FEE PAYMENT PAID IN FULL

Effective date: 20241007

MPN Maintenance fee for patent paid

Free format text: FEE DESCRIPTION TEXT: MF (PATENT, 7TH ANNIV.) - STANDARD

Year of fee payment: 7

U00 Fee paid

Free format text: ST27 STATUS EVENT CODE: A-4-4-U10-U00-U101 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE REQUEST RECEIVED

Effective date: 20250923

U11 Full renewal or maintenance fee paid

Free format text: ST27 STATUS EVENT CODE: A-4-4-U10-U11-U102 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE FEE PAYMENT PAID IN FULL

Effective date: 20250923