Failure recovery scheme for a cloud system
AssigneeEmpire Technology Development LLC
InventorShuichi KURABAYASHI
Life of this patent
Estimating the term…
Abstract
Technologies are generally described for a failure recovery scheme for a cloud system. In some examples, the cloud system may include one or more computing nodes, and one or more network switches configured to relay one or more packets among the one or more computing nodes. A respective one of the computing nodes may include a first processor configured to process the one or more packets to communicate with at least one of the network switches, and a second processor configured to process the one or more packets to communicate with at least one of the other computing nodes in the cloud system.
Claims20
A cloud system comprising:
- one or more computing nodes;
- and
- one or more network switches configured to relay one or more packets among the one or more computing nodes,
- wherein a respective one of the computing nodes includes:
- a first processor configured to process the one or more packets to communicate with at least one of the network switches;
- and
- a second processor configured to process the one or more packets to communicate with at least one of the other computing nodes in the cloud system.
A computing node comprising:
- a first processor configured to process one or more packets to communicate with a network switch;
- and
- a second processor configured to process the one or more packets to communicate with another computing node.
A non-transitory method performed under control of a computing node, comprising:
- detecting a failure in communicating one or more packets with a network switch based on a first communication protocol;
- and
- transferring the one or more packets directly to a target computing node based on a second communication protocol, in response to the detecting of the failure.
Dependencies are inferred from the claim text.