Monitoring network performance and detecting network faults using round trip transmission times

摘要

A computer program product is provided for performing a method including: receiving transmission data over a selected time interval for each of a plurality of communication paths; calculating an average round-trip transmission time for each of the plurality of communication paths over the time interval; comparing an average round-trip transmission time for a communication path having the highest average round-trip transmission time to a threshold value and to a multiple of an average round-trip transmission time for a communication path having the lowest average round-trip transmission time; and determining, based on a result of comparing the highest round-trip transmission time to the threshold value and to a multiple of the lowest round-trip transmission time, whether the time period indicates a delay in communication between the I/O subsystem and the control unit requiring at least one of a monitoring action and a recovery action.

主权项

1. A computer program product for processing communications between a host processor and at least one device connected to the host processor by an input/output processing system, comprising a non-transitory, tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising:
receiving transmission data over a selected time interval for each of a plurality of communication paths between the host processor and a control unit configured to control at least one I/O device; calculating an average round-trip transmission time for each of the plurality of communication paths over the time interval, wherein calculating includes identifying a first communication path having a highest average round-trip transmission time and identifying a second communication path having a lowest average round-trip transmission time; comparing the average round-trip transmission time for the first communication path having the highest average round-trip transmission time to a threshold value; and determining, based on the highest average round-trip transmission time exceeding the threshold value, whether the transmission data indicates a delay in communication between the host processor and the control unit requiring at least one of a monitoring action and a recovery action, wherein determining includes: calculating a ratio of the highest average round-trip transmission time to the lowest average round-trip transmission time; and based on the ratio exceeding a selected value, determining that the transmission data indicates a delay.