Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020193729 - A Networked Computer

Publication Number WO/2020/193729
Publication Date 01.10.2020
International Application No. PCT/EP2020/058619
International Filing Date 26.03.2020
IPC
G06F 15/173 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
15Digital computers in general; Data processing equipment in general
16Combinations 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
163Interprocessor communication
173using an interconnection network, e.g. matrix, shuffle, pyramid, star or snowflake
CPC
G06F 15/173
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
15Digital computers in general
16Combinations 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
163Interprocessor communication
173using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
G06F 15/17318
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
15Digital computers in general
16Combinations 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
163Interprocessor communication
173using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
17306Intercommunication techniques
17318Parallel communications techniques, e.g. gather, scatter, reduce, roadcast, multicast, all to all
G06F 15/17381
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
15Digital computers in general
16Combinations 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
163Interprocessor communication
173using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
17356Indirect interconnection networks
17368non hierarchical topologies
17381Two dimensional, e.g. mesh, torus
G06F 9/3885
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
9Arrangements for program control, e.g. control units
06using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
30Arrangements for executing machine instructions, e.g. instruction decode
38Concurrent instruction execution, e.g. pipeline, look ahead
3885using a plurality of independent parallel functional units
G06F 9/54
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
9Arrangements for program control, e.g. control units
06using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
46Multiprogramming arrangements
54Interprogram communication
G06N 20/10
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
10using kernel methods, e.g. support vector machines [SVM]
Applicants
  • GRAPHCORE LIMITED [GB]/[GB]
Inventors
  • KNOWLES, Simon
Agents
  • DRIVER, Virginia
Priority Data
1904263.927.03.2019GB
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) A Networked Computer
(FR) ORDINATEUR EN RÉSEAU
Abstract
(EN)
A computer comprising a plurality of processing nodes is provided. Each processing node has at least one processor configured to process input data to generate an array of data items. The processing nodes are arranged in cliques in which each processing node of a clique is connected to each other processing node in the clique by first and second clique links. The cliques are inter-connected in rings such that each processing node is a member of a single clique and a single ring. The processing nodes of all cliques are configured to exchange in each exchange step of a machine learning collective via the respective first and second clique links at least two data items with the other processing node(s) in its clique, and all processing nodes are configured to reduce each received data item with the data item in the corresponding position in the array on that processing node.
(FR)
L'invention concerne un ordinateur comportant une pluralité de nœuds de traitement. Chaque nœud de traitement possède au moins un processeur configuré pour traiter des données d'entrée afin de générer un tableau d'éléments de données. Les nœuds de traitement sont disposés en cliques, chaque nœud de traitement d'une clique étant relié à chaque autre nœud de traitement dans la clique par des premières et secondes liaisons de clique. Les cliques sont interconnectées en anneaux de telle façon que chaque nœud de traitement soit membre d'une seule clique et d'un seul anneau. Les nœuds de traitement de toutes les cliques sont configurés pour échanger, lors de chaque étape d'échange d'un collectif d'apprentissage automatique via les premières et secondes liaisons de clique respectives, au moins deux éléments de données avec l'autre ou les autres nœuds de traitement de leur clique, et tous les nœuds de traitement sont configurés pour réduire chaque élément de données reçu avec l'élément de données dans la position correspondante dans le tableau sur le nœud de traitement en question.
Latest bibliographic data on file with the International Bureau