System and method for supporting fast hybrid reconfiguration in a high performance computing environment
AssigneeOracle International Corporation
InventorsEvangelos Tasoulas, Ernst Gunnar Gran, Tor Skeie, Bjørn Dag Johnsen
Life of this patent
Estimating the term…
Abstract
A hybrid reconfiguration scheme can allow for fast partial network reconfiguration with different routing algorithms of choice in different subparts of the network. Partial reconfigurations can be orders of magnitude faster than the initial full configuration, thus making it possible to consider performance-driven reconfigurations in lossless networks.
Claims14
A system for supporting fast hybrid reconfiguration in a high performance computing environment, comprising:
- one or more microprocessors;
- a first subnet, the first subnet comprising
- a plurality of switches arranged in a network architecture comprising a plurality of levels, each of the plurality of levels comprising at least one switch,
- a plurality of end nodes, and
- a management entity, the management entity being selected from the group consisting of:
- a subnet manager;
- a fabric manager;
- and a global fabric manager;
- wherein the plurality of switches is initially configured according to a first configuration method, the first configuration method being associated with a first ordering of the plurality of end nodes;
- wherein a subset of the plurality of switches is configured as a sub-subnet of the first subnet, the sub-subnet of the first subnet comprising a number of levels fewer than the plurality of levels of the first subnet;
- wherein the sub-subnet is reconfigured according to a second configuration method;
- wherein a subset of the plurality of end nodes is associated with the sub-subnet, the subset of the plurality of end nodes belonging to a same partition of the plurality of partitions;
- wherein traffic between the subset of the plurality of end nodes is confined to the subset of the plurality of switches configured as the sub-subnet;
- and
- wherein the second configuration method is associated with a second ordering of at least two of the subset of the plurality of end nodes.
A method for supporting fast hybrid reconfiguration in a high performance computing environment, comprising:
- providing, at one or more microprocessors, a first subnet, the first subnet comprising
- a plurality of switches arranged in a network architecture comprising a plurality of levels, each of the plurality of levels comprising at least one switch,
- a plurality of end nodes, and
- a management entity, the management entity being selected from the group consisting of:
- a subnet manager;
- a fabric manager;
- and a global fabric manager;
- configuring the plurality of switches according to a first configuration method, the first configuration method being associated with a first ordering of the plurality of end nodes;
- configuring a subset of the plurality of switches as a sub-subnet of the first subnet, the sub-subnet of the first subnet comprising a number of levels fewer than the plurality of levels of the first subnet;
- and
- reconfiguring the sub-subnet according to a second configuration method;
- wherein a subset of the plurality of end nodes is associated with the sub-subnet, the subset of the plurality of end nodes belonging to a same partition of the plurality of partitions;
- wherein traffic between the subset of the plurality of end nodes is confined to the subset of the plurality of switches configured as the sub-subnet;
- and
- wherein the second configuration method is associated with a second ordering of at least two of the subset of the plurality of end nodes.
A non-transitory computer readable storage medium, including instructions stored thereon for supporting fast hybrid reconfiguration in a high performance computing environment, which when read and executed by one or more computers cause the one or more computers to perform steps comprising:
- providing, at one or more microprocessors, a first subnet, the first subnet comprising
- a plurality of switches arranged in a network architecture comprising a plurality of levels, each of the plurality of levels comprising at least one switch,
- a plurality of end nodes, and
- a management entity, the management entity being selected from the group consisting of:
- a subnet manager;
- a fabric manager;
- and a global fabric manager;
- configuring the plurality of switches according to a first configuration method, the first configuration method being associated with a first ordering of the plurality of end nodes;
- configuring a subset of the plurality of switches as a sub-subnet of the first subnet, the sub-subnet of the first subnet comprising a number of levels fewer than the plurality of levels of the first subnet;
- and
- reconfiguring the sub-subnet according to a second configuration method;
- wherein a subset of the plurality of end nodes is associated with the sub-subnet, the subset of the plurality of end nodes belonging to a same partition of the plurality of partitions;
- wherein traffic between the subset of the plurality of end nodes is confined to the subset of the plurality of switches configured as the sub-subnet;
- and
- wherein the second configuration method is associated with a second ordering of at least two of the subset of the plurality of end nodes.
Dependencies are inferred from the claim text.