サポートと今すぐチャット
サポートとのチャット

SharePlex 11.4 - Administration Guide

About this Guide Conventions used in this guide Revision History Overview of SharePlex Run SharePlex Run multiple instances of SharePlex Execute commands in sp_ctrl Set SharePlex parameters Configure data replication Configure replication to and from a container database Configure named queues Configure partitioned replication Configure replication to a change history target Configure a replication strategy Configure DDL replication Configure error handling Configure data transformation Configure security features Assign SharePlex users to security groups Start replication on your production systems Monitor SharePlex Prevent and solve replication problems Repair out-of-sync data Tune the Capture process Tune the Post process Recover replication after Oracle failover Make changes to an active replication environment Apply an Oracle application patch or upgrade Back up Oracle data on the source or target Troubleshooting Tips Appendix A: Peer-To-Peer Diagram Appendix B: SharePlex environment variables

Tune the Capture process

This chapter contains instructions for improving the performance of the Capture process to prevent Capture from losing pace with the volume of redo that an Oracle source database generates.

Contents

Disable LOB Mapping

If you have PK/UK logging enabled on the source database (recommended to support more SharePlex features and faster processing), check the setting of the SP_OCT_ENABLE_LOBMAP parameter. This parameter controls whether or not SharePlex uses a LOB map when replicating tables that contain out-of-row LOB columns. The LOB map is used by the Capture process to map LOBIDs and rows when PK/UK logging is not enabled. LOB mapping is enabled by default. The SHAREPLEX_LOBMAP table stores these mappings.Transactions with numerous LOB operations can slow down Capture because it needs to maintain and refer to the mappings. If PK/UK logging is enabled on the database, you can disable LOB mapping by setting this parameter to 0.

To disable LOB mapping during active replication:

  1. Run sp_ctrl on the source system.
  2. Set SP_OCT_ENABLE_LOBMAP to 0.

    sp_ctrl> set param SP_OCT_ENABLE_LOBMAP 0

  3. Stop Capture.

    sp_ctrl> stop capture

  4. Truncate the SHAREPLEX_LOBMAP table.
  5. Restart Capture.

    sp_ctrl> start capture

Tune Capture on Exadata

The Capture process can be configured to use multiple capture threads for faster performance on an Exadata system. Capture reads directly from the logs on the Exadata ASM disks.

The SP_OCT_ASM_MULTI_OCI parameter controls the number of threads that Capture uses to read the redo logs.

The value for this parameter must be set to at least 2 but no more than the number of disks in the redo log disk group.

A large number of threads is not required, and performance actually diminishes with too many threads. The more threads, the more memory Capture requires. Start with a small number of threads and monitor performance, then add threads if needed until you obtain an ideal balance between performance gain and memory usage.

To configure SharePlex for multi-threaded capture on Exadata:

  1. Run sp_ctrl.
  2. Set the SP_OCT_ASM_MULTI_OCI parameter to the number of threads that you want Capture to use.

    sp_ctrl> set param SP_OCT_ASM_MULTI_OCI 3

  3. Restart Capture.

Note: Capture automatically adjusts its buffer size to the value of the AU_SIZE parameter that is set for the disk group where the logs reside. This is the recommended buffer size for best performance and should not be changed. The SP_OCT_ASM_MULTI_OCI_BLOCK_SIZE parameter can override the default behavior if necessary.

Tune Checkpointing

Capture checkpoints it state to disk on a regular basis to support recovery. This information includes the log and location within that log of the most recently processes data. In a database environment where there are frequent log switches, a switch can occur before SharePlex writes its checkpoint. You can use the SP_OCT_CHECKPOINT_LOG parameter to ensure that Capture issues a checkpoint before a log switch.

The checkpoint is triggered when Capture lags a specified number of logs behind Oracle. For example, with the default of 2, Capture does a checkpoint when it falls 2 or more logs behind Oracle.

The range of permissible values for this parameter is from 2 (the default) to a value equal to the number of logs you are using. A value of 0 disables this feature.

関連ドキュメント

The document was helpful.

評価を選択

I easily found the information I needed.

評価を選択