A simple, flexible and complete out-of-the-box solution
The PureData System for Operational Analytics is designed to provide very high throughput and concurrency. Each data module has more than 6 GB/sec of sustained I/O bandwidth, and as data modules are added, bandwidth increases proportionately. With the underlying strengths of IBM DB2 software, the PureData System is designed to handle 1000+ concurrent operational queries1. Overall, performance for complex queries is 29% faster than the previous release of PureData for Operational Analytics4.
The system is a complete, out-of-the-box solution for operational analytics that provides both the simplicity of an appliance and the flexibility of a custom solution.
The PureData System Control Console is the single application used to support all management and monitoring of the system ensuring efficient performance. It provides a singular monitoring and management tool to administer system wide updates and obtain a view into the current state of the system. Installation services, one-call support and hardware and software maintenance are also available.
A modular, flexible system design allows organizations to acquire the PureData System at the correct size for their current needs, a choice of six configurations is available, and scale incrementally to petascale capacity2.
With in-database analytics, you run analytics on your data where it resides—in the warehouse. This eliminates the time, cost and risk associated with copying data out of the warehouse to analyze it.
Using Adaptive Compression features, PureData System for Operational Analytics can automatically compress indexes and temporary tables to help reduce storage costs. Data row compression contributes to storage space savings and helps reduce I/O overhead—and the stored pages are also compressed, which further enhances the compression on disk. Clients have experienced cases of 10x storage space savings with Adaptive Compression3.
PureData for Operational Analytics supports the concept of tiered storage. According to this concept, active or frequently used data resides on faster flash storage while the inactive/cool data is placed on slower spinning disks. Not only does this help improve the query performance, the total cost of storage is also reduced using this hybrid tiered storage system (as compared to an all flash storage system).
1Based on IBM internal tests of prior generation system, and on system design for normal operation under expected typical workload. Individual results may vary.
2Total raw data capacity based on XX Large configuration with five full rack data expansion add-ons.
3Based on client testing in the DB2 10 Early Access Program.
4As compared to previous generation of PureData for Operational Analytics, using the same DB2 parameters and settings. Individual results may vary.