SBR Carrier provides SSR cluster performance enhancements, as well as other product improvements. However, depending on the current release of software you are running, and which upgrade method you use to perform the upgrade, you may or may not be able to take advantage of the performance enhancements.
The choice of which upgrade method you use depends on your tolerance for downtime in the SSR cluster. You can use one of the following two methods to upgrade your SSR cluster:
Rolling restart upgrade—Use this method to upgrade existing SBR Carrier release SSR clusters to Release 8.5.0. This upgrade provides numerous product improvements and corrections, but does not provide performance enhancements to the cluster. To experience product improvements and corrections, as well as improved performance in your cluster, you must upgrade using the backup, destroy, and re-create upgrade method. For complete details on the changes to the Release 8.5.0 software, see the SBRC Release 8.5.0 Release Notes. When using this upgrade method, we recommend you mitigate the risk of downtime in your cluster by using a transition server; for more information see Using a Transition Server to Mitigate Downtime While Upgrading Your Cluster.
Backup, destroy, and re-create your cluster—Use this method to upgrade from clusters running previous releases of SBR Carrier software to Release 8.5.0. This upgrade method results in the data nodes using a multi-threaded process which improves performance over the single-threaded data node processes used in previous SBR Carrier software releases. In addition, Release 8.5.0 provides many improvements and corrections to the SBRC software. For complete details on the changes to Release 8.5.0 software, see the SBRC Release 8.5.0 Release Notes. Using the backup, destroy, and re-create upgrade method requires cluster downtime. For this reason, we recommend you use a transition server during the upgrade process. For details on using a transition server, see Using a Transition Server to Mitigate Downtime While Upgrading Your Cluster.