|System z on Facebook
What you did not know about Big Data on System z
Pratin Ashtekar 2700046SAG firstname.lastname@example.org | | Tags:  big_data ibmiod bao economics db2 z/os mark_simmonds iod2013 system_z_software | 1,880 Visits
By Mark Simmonds,
Sr. Product Marketing Manager - Big data for System z .
Last time I blogged here I talked about the impact of security on big data. Just a reminder... "big data" is ALL data, from the rich, highly structured data in databases like DB2 for z/OS and IMS and data warehouses to sensor-to-sensor machine generated data, emails, social media sites, documents and beyond. In short any data relevant to your organization that can help deliver better business outcomes and competitive advantage.
Big data requires integration and governance (my relevant post Big data - past, present and future ) to succeed . So with the recent announcement of DB2 11 for z/OS and IMS 13, things just got a whole lot more interesting - and easier - and faster ... for example IMS blows through the 100,000 trans/sec barrier. Now you can invoke HDFS queries from within your SQL, integrating those result sets back into your databases or warehouse to augment your understanding of what is happening in your business through deeper insights.
But why bring that data onto the mainframe? Think about it.... Organizations have made huge investments in their computing platforms that hold those mission critical transaction data sources. These have become the foundation of today's modern analytics so much so that many have co-located their data warehouses with this data to reduce latency of updates, make best use of scarce resources, enjoy the same high qualities of service of the transactional data - and so on. So business analytics applications like Cognos BI, SPSS predictive analytics, QMFcan also enjoy those same qualities of service - even if you're accessing them from your mobile apps. For the majority of the time the industry has been focused on the "bigness" of big data - particularly volume and velocity factors. The IBM DB2 Analytics Accelerator for z/OS v 4.1nails those factors. Plug it in, experience value in just a couple of hours and watch queries that took half a day to run finish in seconds - and no need for any user intervention.
Want to read more? I invite you to register for these two great white papers :