Method for deploying a task in a supercomputer, method for implementing a task in a supercomputer, corresponding computer program and supercomputer

1. A method for deploying a task in a computer cluster forming a supercomputer comprising:

computers each having at least one network interface forming a node; and

a network for interconnecting nodes, comprising:

switches each having network interfaces, and

links each connecting either a node and a network interface of one switch or two network interfaces of two switches respectively;

the method comprising:

allocating nodes to the task to form allocated nodes;

determining in the network, a subnetwork, for interconnecting the allocated nodes, satisfying one or more predefined determination criteria comprising a first criterion whereby the subnetwork uses only links that are not allocated to any other task already deployed or that are allocated to fewer than N other tasks already deployed, where N is a predefined number equal to one or more;

allocating the subnetwork, and the links belonging to said subnetwork, to the task, to form an allocated subnetwork; and

implementing inter-node communication routes in the allocated subnetwork.

2. A method according to claim 1 , wherein the predefined number, N, is one.

3. A method according to claim 1 , wherein the one or more predefined determination criteria for determining the subnetwork further include a second criterion according to which the subnetwork, among at least two subnetworks satisfying the one or more predefined determination criteria other than the second criterion, uses a largest number of switches already allocated each to at least one task already deployed.

4. A method according to claim 3 , wherein determining the subnetwork comprises determining the at least two subnetworks satisfying the one or more predefined determination criteria other than the second criterion, and then selecting from among the at least two subnetworks a subnetwork that uses a largest number of switches each already allocated to at least one task already deployed.

5. A method for implementing a task, comprising:

deploying the task in a supercomputer, in accordance with a deployment method according to claim 1 ;

executing the task; and

at an end of the execution of the task:

detecting at least one link that is not allocated to any task, and

setting each detected link in an idle state, wherein the each detected link requires less energy consumption than a link associated with at least one task.

6. A method according to claim 5 , comprising, for setting a link in the idle state, switching off two network interfaces that said link connects so that no current or light flows through said link.

7. A method according to claim 5 , comprising, in order to set a link in the idle state, reducing an operating frequency of at least one of two network interfaces that said link connects.

8. A non-transitory computer readable including a computer program comprising instructions for executing steps of a method according to any claim 1 , when said computer program is executed on one or more computers.

9. A computer cluster forming a supercomputer comprising:

computers each having at least one network interface forming a node;

a network for interconnecting nodes, comprising:

switches each having network interfaces, and

links each connecting either a node and a network interface of one switch, or two network interfaces of two respective switches;

the supercomputer further comprising software and/or hardware means for deploying a task, designed to implement a step of:

allocating nodes to the task to form allocated nodes;

the supercomputer comprising software and/or hardware means for deploying a task, designed to implement steps of:

determining, in the network, a subnetwork, for interconnecting the allocated nodes, satisfying one or more predefined determination criteria comprising a first criterion whereby the subnetwork uses only links that are not allocated to any other task already deployed or that are allocated to fewer than N other tasks already deployed, where N is a predefined number equal to one or more;

allocating the subnetwork, and the links belonging to said subnetwork, to the task, to form an allocated subnetwork; and

implementing inter-node communication routes in the allocated subnetwork.