The data warehouse remains an a key part of any data architecture, but as always, the challenge is in managing a growing volume of data types and sources and responding quickly to new needs for analytics. This can often result in using the best database or data warehouse for data types. Think about the tradeoffs between data, applications and workloads such as Hadoop vs. structured data; web clickstreams vs. financial data; marketing insights vs. quarterly reporting.
dashDB Local, part of the IBM Hybrid Data Architecture
To respond to all these needs you need multiple solutions. IBM calls this a Hybrid Data Architecture and offers solutions for different types of data and workloads that share common features. IBM dashDB Local is one of these platforms. It is a configured data warehouse that is delivered via Docker containers -- so you can quickly stand up a new data warehouse in a matter of minutes on an on-premises server, private cloud or even public cloud -- virtually any platform that supports Docker containers.
Now available on IBM Power Architecture
dashDB Local has been available on Intel servers and the news is that it is now available on the IBM Power Architecture. This is significant because the Power Systems platform is optimized for big data and analytics workloads. A dashDB Local container can run transparently on Power Systems and is optimized to take advantage of the unique features of these systems. Whether you use dashDB on Intel or Power Systems, you fill find the same user, database, OS management, monitoring and integration experience across these systems. A benchmark has found that dashDB Local has 1.2x more throughput on Power Systems with fewer cores compared to Intel systems with more cores.