Processing

Please wait...

Settings

Settings

Goto Application

1. WO2019042312 - DISTRIBUTED COMPUTING SYSTEM, DATA TRANSMISSION METHOD AND DEVICE IN DISTRIBUTED COMPUTING SYSTEM

Publication Number WO/2019/042312
Publication Date 07.03.2019
International Application No. PCT/CN2018/102919
International Filing Date 29.08.2018
IPC
G06F 9/48 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR 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
48Program initiating; Program switching, e.g. by interrupt
CPC
G06F 16/24568
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
20of structured data, e.g. relational data
24Querying
245Query processing
2455Query execution
24568Data stream processing; Continuous queries
G06F 16/9024
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
90Details of database functions independent of the retrieved data types
901Indexing; Data structures therefor; Storage structures
9024Graphs; Linked lists
G06F 9/48
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
48Program initiating; Program switching, e.g. by interrupt
G06F 9/4806
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
48Program initiating; Program switching, e.g. by interrupt
4806Task transfer initiation or dispatching
G06F 9/5061
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
50Allocation of resources, e.g. of the central processing unit [CPU]
5061Partitioning or combining of resources
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
Applicants
  • 华为技术有限公司 HUAWEI TECHNOLOGIES CO., LTD. [CN]/[CN]
Inventors
  • 林健 LIN, Jian
  • 夏命榛 XIA, Mingzhen
Priority Data
201710769632.831.08.2017CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) DISTRIBUTED COMPUTING SYSTEM, DATA TRANSMISSION METHOD AND DEVICE IN DISTRIBUTED COMPUTING SYSTEM
(FR) SYSTÈME INFORMATIQUE RÉPARTI, PROCÉDÉ ET DISPOSITIF DE TRANSMISSION DE DONNÉES DANS UN SYSTÈME INFORMATIQUE RÉPARTI
(ZH) 分布式计算系统,分布式计算系统中数据传输方法和装置
Abstract
(EN)
A distributed computing system is disclosed. A first computing node and a second computing node in the distributed computing system both have stored therein information relating to a name, size, and communication peer identifier of a first data flow diagram parameter in a data flow diagram, and the first data flow diagram parameter is stored in the first computing node. The first computing node and the second computing node use the same interface parameter saved in the respective nodes thereof to generate algorithms and the above information of the first data flow diagram parameter, thereby generating respective triplets thereof that are used as interface parameters of a messaging interface MPI primitive for transmitting the first data flow diagram parameter between the first computing node and the second computing node. In this way, the first computing node and the second computing node can transmit the data flow diagram parameter using the MPI primitive without negotiation, thereby improving data communication efficiency between computing nodes and improving computing efficiency of the distributed computing system with respect to the data flow diagram.
(FR)
L'invention concerne un système informatique réparti. Un premier nœud informatique et un second nœud informatique dans le système informatique réparti renferment tous deux, stockées dans ceux-ci, des informations relatives à un nom, une taille, et un identifiant d'homologue de communication d'un premier paramètre de diagramme de flux de données dans un diagramme de flux de données, et le premier paramètre de diagramme de flux de données est stocké dans le premier nœud informatique. Le premier nœud informatique et le second nœud informatique utilisent le même paramètre d'interface sauvegardé dans leurs nœuds respectifs pour générer des algorithmes et les informations ci-dessus du premier paramètre de diagramme de flux de données, générant ainsi des triplets respectifs de celles-ci qui sont utilisés en tant que paramètres d'interface d'une primitive MPI d'interface de messagerie pour transmettre le premier paramètre de diagramme de flux de données entre le premier nœud informatique et le second nœud informatique. De cette façon, le premier nœud informatique et le second nœud informatique peuvent transmettre le paramètre de diagramme de flux de données à l'aide de la primitive de MPI sans négociation, améliorant ainsi le rendement de communication de données entre des nœuds informatiques et améliorant ainsi le rendement de calcul du système informatique réparti par rapport au diagramme de flux de données.
(ZH)
一种分布式计算系统,该分布式计算系统中的第一计算节点和第二计算节点都保存有数据流图中第一数据流图参数的名称、大小和通信对端标识的信息,第一计算节点中保存有该第一数据流图参数,第一计算节点和第二计算节点使用各自节点中存储的相同的接口参数生成算法和第一数据流图参数的上述信息,生成各自的三元组,该三元组被用作第一计算节点和第二计算节点间用于传输第一数据流图参数的消息传递接口MPI原语的接口参数。这样,第一计算节点与第二计算节点就可以无需协商地使用MPI原语传输数据流图参数,能够提高计算节点间数据的通信效率,从而提升分布式计算系统对数据流图的计算效率。
Also published as
Latest bibliographic data on file with the International Bureau