IP Library › Granted Patent US 10,826,779
Granted Patent B2
US 10,826,779 · App. 15/534,816 · Granted Nov 3, 2020

Method for managing a network of compute nodes

Inventor: Matthieu Perotin (Grenoble, FR)
Assignee: BULL SAS
H04L41/12G06F9/5083G06F15/16G06F15/161G06F15/17362H04L67/10
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 10,826,779
App. No.
15/534,816
Granted
Nov 3, 2020
Kind
B2
Abstract

A method of managing a network of calculation nodes interconnected by a plurality of interconnection devices, includes organizing the calculation nodes into groups of calculation nodes, for each group of calculation nodes, connecting the interconnection devices interconnecting the nodes of the group to a group management node, the management node being dedicated to the group of calculation nodes on each management node execution of an administration function by the implementation of independent management modules, each management module of a management node being able to communicate with the other management modules of the same management node.

Claims (35)

1. A method for managing a network of compute nodes interconnected by a plurality of interconnection devices, the method comprising:

organizing compute nodes into groups of compute nodes;

for each group of compute nodes, connecting interconnection devices interconnecting the compute nodes of the group to a group management node, wherein the group management node is dedicated to the group of compute nodes,

in each group management node, executing a management function through an implementation of independent management modules, wherein each management module for managing a management node is configured to communicate with other management modules of the same management node, wherein the communication between the management modules of the same management node is asynchronous, and wherein at least one group management node is connected to a level-2 management node, and

determining, by the at least one group management node, whether to transmit one or more messages to the level-2 management node based on processing of the one or more messages by the at least one group management node.

2. The method according to claim 1 , wherein the group management node is connected in an out-of-band mode to the interconnection devices.

3. The method according to claim 1 , wherein the management modules include:

a network topology management module,

a routing management module,

a network supervision module, and

a network performance measurement module.

4. The method according to claim 1 , wherein communications between the at least one group management node and the level-2 management node are accomplished in an out-of-band mode.

5. The method according to claim 1 , wherein a management node module processes the one or more messages received from at least one interconnection device to reduce quantity of data transmitted to the level-2 management node.

6. The method according to claim 3 , wherein the network topology management module is configured to maintain up-to-date a topology database and a status of each connector of each interconnection device.

7. The method according to claim 3 , wherein the routing management module is configured to calculate routing tables.

8. The method according to claim 3 , wherein the network supervision module is configured to manage messages transmitted by the interconnection devices and aggregate data produced by other management modules.

9. The method according to claim 8 , wherein the data is aggregated before transmitting the aggregated data to a level-2 management node.

10. The method according to claim 3 , wherein the network performance measurement module is configured to obtain and record data concerning a performance of the interconnection devices.

11. The method according to claim 10 , wherein the data concerning the performance of the interconnection devices includes a transmission rate in each port, number of packets, length of queues, number of errors, and protocols used.

12. The method of claim 1 , wherein the level-2 management node is connected to the at least one group management node via a switch.

13. The method of claim 1 , wherein the processing of the one or more messages includes determining a number of the one or more messages, the method further comprising:

determining, by the at least one group management node, not to transmit the one or more messages to the level-2 management node when the number of the one or more messages exceeds a predetermined threshold.

14. The method of claim 13 , wherein the one or more messages are received by the at least one group management node from one or more of the interconnection devices.

15. The method of claim 14 , wherein the level-2 management node is dedicated to a group of a plurality of group management nodes.

16. A non-transitory machine readable storage medium storing computer program instructions that, when executed, by a computer cause the computer to perform operations comprising:

executing, for a management node, a management function through an implementation of independent management modules;

communicating, by a management module of the management node, with other management modules of the same management node,

wherein the communication between the management module and the other management modules of the same management node is asynchronous,

wherein the management node is connected to a group of compute nodes via interconnection devices, and

wherein the management node is connected to a level-2 management node; and

determining by the management node, whether to transmit one or more messages to the level-2 management node based on processing of the one or more messages by the management node.

17. A system comprising:

a group of compute nodes;

interconnection devices connecting the group of compute nodes to a management node, wherein the management node is dedicated to the group of computer nodes; and

the management node, wherein the management node is configured to execute a management function through an implementation of independent management modules, wherein a management module for managing the management node is configured to communicate with other management modules of the same management node, wherein the communication between the management modules of the same management node is asynchronous, wherein the management node is connected to a level-2 management node, and wherein the management node determines whether to transmit one or more message to the level-2 management node based on processing of the one or more messages by the management node.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 18, 2017
From: PEROTIN, MATTHIEU
To: BULL SAS
Reel/Frame 043895/0209 →
Priority Claims (1)
FR 14 62157 · Dec 10, 2014 · national
Continuity (1)
Related Publication 20170353358A1 · Dec 7, 2017