Chapter 11. IMS Connect Extensions 193
Figure 11-20 Candidate list for a single transaction
Step 1 in the diagram shows the incoming message request. Step 2 shows that TX1 has a
transaction affinity of datastores IMP1, IMT1, and IMD1. Step 3 shows that IMD2 has a
datastore affinity of datastores IMD1 and IMD2. Step 4 shows the intersection of transaction
affinity and datastore affinity as IMD1. The message is routed from IMD2 to the new target
datastore of IMD1.
11.3.2 Workload balancing
Workload balancing allows IMS Connect Extensions to redirect incoming transaction
requests to one of multiple IMS datastores, thereby balancing the transaction workload
across multiple IMS systems, ensuring both availability and responsiveness for the remote
client.
Workload balancing uses a weighted rotate algorithm that considers the processing capacity
of a datastore. Datastores with greater capacity receive more transactions. Basic rotation is
achieved by assigning each datastore an equal weighting value. You specify the relative
weighting of each datastore during the datastore definition process. See Figure 11-19 on
page 192. The Capacity weight rating field accepts values in the range of 1 to 100.
If the workload balancing option on the system definition is set to Active and IMS Connect
Extensions identifies that a datastore is no longer available, the datastore is not used to
process incoming message requests.
Figure 11-21 on page 194 illustrates workload balancing. The figure shows an IMS system
containing eight datastores and assumes an incoming message request running TX1 on
datastore IMD1 and that the definitions for both TX1 and IMD1 have transaction routing set to
Active.

Get IMS Connectivity in an On Demand Environment: A Practical Guide to IMS Connectivity now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.