Cluster Wide Consistent Detection of Interconnect Failures

At least one non-transitory computer readable medium storing a computer program product for propagating a detection of a failure of a link to each of a plurality of nodes of a cluster, the computer program product comprising:

• program code for detecting a link-down event comprising a failure of a specific link of a specific node of the cluster;
• and
• program code for disseminating a notification of the failure of the specific link to the plurality of nodes of the cluster, such that the nodes process the link-down event in parallel;
• wherein the disseminating of the notification of the failure of the specific link is executed such that the nodes of the cluster receive the notification prior to learning of the link-down event from expiration of corresponding heartbeats.

A computer system for propagating a detection of a failure of a link to each of a plurality of nodes of a cluster, the computer system comprising:

• a processor;
• computer memory;
• a link failure detecting module residing in the computer memory, for detecting a link-down event comprising a failure of a specific link of a specific node of the cluster;
• and
• a broadcasting module residing in the computer memory, for broadcasting a message notifying the nodes of the cluster of the detected link failure, such that the nodes of the cluster receive the notification prior to learning of the link-down event from expiration of corresponding heartbeats, and process the link-down event in parallel.