Originally posted @ Gigaom Research 9/15/2014
IBM’s BLU Acceleration is starting to make inroads in addressing common issues faced by enterprise IT organizations. BLU Acceleration is IBM’s in-memory database processing is based on DB2. In-memory processing provides significant performance advantages over traditional methods by eliminating the need to constantly read and write from hard drives. Instead, BLU Acceleration pulls the data into memory and performs the analytical processes within memory. This results in accelerated performance during analytical operations.
IBM’s Cancun Release of BLU Acceleration includes a number of new enhancements to the core product including support for Shadow Tables, Oracle & SAP and IBM’s POWER8 integration.
The Cancun Release provides the ability to conduct operations using tables on top of existing tables within the database. Interestingly, the Shadow Tables are column-oriented versus traditional row-oriented tables. This provides a second layer of capability without impacting the performance of core table structures.
Integration with Oracle and SAP
In different ways, IBM is providing greater integration with both Oracle and SAP. For Oracle, IBM is providing tools to migrate existing Oracle environments into IBM DB2 in order to retain existing work. This is key when considering a move from one database platform to another.
For SAP, IBM has provided greater integration with SAP BW (Business Warehouse) beyond the core offering. In doing so, SAP workloads can leverage BLU Acceleration with a number of enhancements. It will be interesting to see how this space evolves compared with SAP’s own in-memory solution, SAP HANA.
Integration with IBM POWER8 processor
The last of the Cancun Release enhancements provides optimizations that fully leverage the POWER8 processor such as concurrent multithreading and 128-bit register instructions. In doing so, IBM creates a powerful combination of tying software applications to engage specific hardware features and therefore optimizing the application for the hardware platform.
IBM BLU Acceleration provides a number of opportunities for in-memory database process. The catch is that it leverages IBM’s DB2 database. In order to leverage the benefits, customers will need to migrate from their existing platform (Oracle, SQL, etc) to DB2.
For organizations looking to enhanced performance in their Big Data and Analytics operations, in-memory processing is key. DB2 with BLU Acceleration is an interesting solution but may pose a moment of pause for non-DB2 customers when considering the requirements to migrate. For Oracle customers, the migration tools in the Cancun Release will help with this process.