Performance Improvements Notable in Latest IDAA Release
Here's a quick list of other key enhancements:
* Query prioritization -- In release 2.1 you could prioritize a mixed workload and give more resources to a given user or a given set of processes. While these actions were honored in DB2 through the workload manager, when the queries got down to the IDAA, the prioritization rules weren't enforced. However, with this release the workload management is applied to the queries that are routed to the IDAA. This will ensure that the most important work is completed first.
* Incremental updates -- With version 3.1, updates to local DB2 tables are continually synchronized with tables in the accelerator. This feature supports applications that have a need for high data currency, such as operational business intelligence (BI).
* High capacity -- With version 3.1, the Netezza 1000 scales from one to 10 cabinets. This is designed to support the use of larger accelerators based on the larger data warehouses found in DB2 on System z.
* Unload Lite -- This utility extracts data from DB2 and writes it to a USS pipe, where it is then transferred via TCPIP to the IDAA. With Unload Lite data conversions are handled on the IDAA rather than within z/OS, thus reducing costs.
* SAP plug-in -- Customers running SAP with DB2 on z/OS will be happy to learn that an SAP plug-in provides a natural extension to SAP NetWeaver BW. This is designed to not only improve performane, but reduce processing costs by freeing up MIPS and storage space used for indexing.
The IDAA supports the current z196 and z114 series of System z as well as the new next generation System z zEnterprise EC 12 (zEC12). The performance gains available with IDAA on the zEC12 are amazing. Read the release notes for details.
If you're using IDAA 3.1, please share your thoughts in Comments.
Note: A quick follow-up on last week's post. To my great relief, PlanetDB2.com is back. Apparently a technical glitch caused the site to aggregate posts meant for PlanetBigData.com.