Background copy bandwidth impact on foreground I/O latency

The background copy bandwidth determines the rate at which the background copy for Metro Mirror or Global Mirror Copy Services are attempted.

The background copy bandwidth can affect foreground I/O latency in one of three ways:
  • If the background copy bandwidth is set too high for the intercluster link capacity, the following results can occur:
    • The background copy I/Os can back up on the intercluster link
    • For Metro Mirror, there is a delay in the synchronous secondary writes of foreground I/Os
    • For Global Mirror, the work is backlogged, which delays the processing of writes and causes the relationship to stop
    • The foreground I/O latency increases as detected by applications
  • If the background copy bandwidth is set too high for the storage at the primary site, background copy read I/Os overload the primary storage and delay foreground I/Os.
  • If the background copy bandwidth is set too high for the storage at the secondary site, background copy writes at the secondary overload the secondary storage and again delay the synchronous secondary writes of foreground I/Os.
    • For Global Mirror, the work is backlogged and again the relationship is stopped

To set the background copy bandwidth optimally, you must consider all three resources (the primary storage, the intercluster link bandwidth, and the secondary storage). Provision the most restrictive of these three resources between the background copy bandwidth and the peak foreground I/O workload. You must also consider concurrent host I/O because if other write operations arrive at the primary cluster for copy to the remote site, these write operations can be delayed by a high level of background copy and the hosts at the primary site receive poor write-operation response times.

This provisioning can be done by the calculation above or by determining how much background copy can be allowed before the foreground I/O latency becomes unacceptable and then backing off to allow for peaks in workload and some safety margin.

Example

If the bandwidth setting at the primary site for the secondary cluster is set to 200 MBps (megabytes per second) and the relationships are not synchronized, the SAN Volume Controller attempts to resynchronize the relationships at a maximum rate of 200 MBps with a 25 MBps restriction for each individual relationship. The SAN Volume Controller cannot resynchronize the relationship if the throughput is restricted. The following can restrict throughput:
  • The read response time of backend storage at the primary cluster
  • The write response time of the backend storage at the secondary site
  • Intercluster link latency
Library | Support | Terms of use | Feedback
© Copyright IBM Corporation 2003, 2009. All Rights Reserved.