CONGESTION CAUSATION IN A NETWORK INTERCONNECT

摘要

A method and system for detecting congestion in a network of nodes, abating the network congestion, and identifying the cause of the network congestion is provided. A congestion detection system may comprise a detection system, an abatement system, and a causation system. The detection system monitors the performance of network components such as the network interface controllers and tiles of routers to determine whether the network is congested such that a delay in delivering packets becomes unacceptable. Upon detecting that the network is congested, an abatement system abates the congestion by limiting the rate at which packets are injected into the network from the nodes. Upon detecting that the network is congested, a causation system may identify the job that is executing on a node that is the cause of the network congestion.

主权项

1. A method performed by a computer system for identifying a job executing on a node, the job being a candidate for causing congestion in a network of nodes connected via connection devices, each node connected to a local controller, each of a plurality of the local controllers connected to a supervisory controller via an out-of-band network connection, the method comprising:
determining an ejection rate of packets from the network to each node based on information collected by the local controllers during abatement; identifying at the supervisory controller a node as a candidate node for executing the job causing the congestion in the network based on the ejection rates of packets from the network to the nodes; identifying a job executing on the candidate node; and indicating that the identified job is a candidate job for causing the network congestion.