GLVM DR Sizing tool
The performance of a volume group in an asynchronous Geographical Logical Volume Manager (GLVM) is affected by the bandwidth between sites and the cache configuration of a logical volume. For system requirement updates, you must monitor the network and performance of data replication of the GLVM volume group.
In PowerHA® SystemMirror® Version 7.2.7, the GLVM disaster recovery (DR) Sizing tool is used to monitor data replication and to provide suitable recommendations.
- Collects the data writes and cache information for the local and remote sites.
- Analyzes the collected data.
- Provides recommendations to optimize the cache size and the network bandwidth.
- Creates a server on the home node, with all the remote nodes connecting to this server.
- Transfer data packets on the network and collect network performance statistics.
- Provide information about network bandwidth and network latency.
The GLVM DR Sizing tool includes the cl_survey utility that is used to evaluate the current disk usage on the system. The cl_survey utility provides information to specific network that has sufficient bandwidth that can be used in an asynchronous GLVM volume group. The cl_survey utility works similar to UNIX performance monitoring programs (for example, the clperfstat program). For more information, see the cl_survey command.
- An asynchronous GLVM must be configured in PowerHA SystemMirror, and the cluster services must be in the STABLE state.
- You cannot run the tool multiple times.
- Raw logical volumes are not supported.
- You must not change the configuration when the cl_survey tool is running. The cl_survey tool might stop if it detects any configuration changes when it is running.