IBM Support

Collocation and number of concurrent migration processes

Technote (troubleshooting)


Problem(Abstract)

This document provides information on the number of concurrent migration process that can run concurrently when collocation is enabled.

Resolving the problem

The number of concurrent migration processes that can run when collocation is enabled depends on what data is available to be migrated from the source storage pool and the type of collocation used in the target storage pool.
In the examples below, the source storage pool is named DISKPOOL and migrates to the target tape pool named TAPEPOOL. The DISKPOOL storage pool is defined with "Migration Processes=4"

  1. The TAPEPOOL storage pool is defined with collocation=node and the DISKPOOL storage pool contains backup data from six (6) different nodes.
    Result : The migration from DISKPOOL to TAPEPOOL will use four (4) migration processes, one for each node, up to the maximum of 4 processes configured.
  2. The TAPEPOOL storage pool is defined with collocation=node and the DISKPOOL storage pool contains backup data from two (2) different nodes.
    Result :The migration from DISKPOOL to TAPEPOOL will use two (2) migration processes, one for each node.
  3. The TAPEPOOL storage pool is defined with collocation=filespace and the DISKPOOL storage pool contains backup data from two (2) different nodes. Backup data for NODE1 includes data from two (2) different filespaces, C: and D:, and NODE2 includes data from filespace E:.
    Result : The migration from DISKPOOL to TAPEPOOL will use three (3) migration processes. one for each filespace.
  4. The TAPEPOOL storage pool is defined with collocation=group and the DISKPOOL storage pool contains backup data from eight (8) different nodes and all eight (8) nodes belong to the same nodegroup.
    Result :The migration from DISKPOOL to TAPEPOOL will use one (1) migration process, one for the nodegroup.
  5. The TAPEPOOL storage pool is defined with collocation=group and the DISKPOOL storage pool contains backup data from eight (8) different nodes. Three (3) nodes belong to one nodegroup and five (5) nodes belong to another nodegroup.
    Result :The migration from DISKPOOL to TAPEPOOL will use two (2) migration processes, one for each nodegroup
  6. The TAPEPOOL storage pool is defined with collocation=group and the DISKPOOL storage pool contains backup data from eight (8) different nodes. Three (3) nodes belong to one nodegroup, four (4) nodes belong to another nodegroup and one node does not belong to any group..
    Result :The migration from DISKPOOL to TAPEPOOL will use three (3) migration processes, one for each nodegroup and one for the node that does not belong to any nodegroup.
  7. The TAPEPOOL storage pool is defined with collocation=group and the DISKPOOL storage pool contains backup data from one (1) single node which has five (5) filespaces. Two (2) filespaces belong to one file space collocation group (new with Tivoli Storage Manager V7.1) and thress (3) filespaces belong to another file space collocation group.
    Result :The migration from DISKPOOL to TAPEPOOL will use two (2) migration process, one for each file space collocation group

Related information

How the server selects files to migrate
Specifying multiple concurrent migration processes

Document information

More support for: Tivoli Storage Manager
Server

Software version: All Supported Versions

Operating system(s): Platform Independent

Reference #: 1636484

Modified date: 24 February 2017


Translate this page: