The architecture of the cluster hardware is an important consideration when planning a deployment. The type of workload is important: consider whether the planned cluster usage will be CPU-intensive, I/O-intensive, or memory-intensive. Hardware recommendations and cluster configuration vary by use case. Think about how data will be loaded into and out of the cluster, and how much data is likely to be transmitted over the network.

Planning a cluster often involves tuning key ratios, such as:

• Disk I/O speed to CPU processing power
• Storage capacity to network speed
• Number of nodes to network speed

The new IBM’s cutting edge technology storage systems were implemented as a core storage cluster for entire Bank’s SAN infrastructure. During implementation zero downtime migration of existing storage was performed – Flex Cube Core Banking, Flex Cube Direct Banking, Datawarehouse, BI, Business Support Systems, Integration Bus, BPM and etc. All existing storage was totally reused as in backend under IBM Spectrum Virtualize which provided bank with single source of storage access to simplify management and kept an outstanding level of investment protection made in the past in old infrastructure.

Nano-Latency NVM FlashCore technology is used to accelerate entire Bank’s storage infrastructure. IBM HyperSwap and clustering ensured 99.9999% storage availability and significantly reduced risks of data access outages and in place capacity availability. Many innovative technologies, such as Tiering, Compression and Data Encryption for Virtualized devices were implemented and others. After implementation overall performance of storage systems increased over 5 times ensuring absolute data availability combined with cost effective, quick and easy scalability.

Cluster of IBM Storwize V7000 Gen-3 Model 724

IBM Spectrum Virtualize