z Systems for Analytics Data—You Move, You Lose
Solutions are available to help save MIPS, money
5/20/2015 12:30:23 AM |
By Avijit Chatterjee
Analytics on our latest generation IBM z13 mainframe running DB2 11 for z/OS with IBM DB2 Analytics Accelerator beats the biggest competitor, on a comparable platform, by two times performance and four times price performance (using street prices). Yet, customers have been known to move data off the mainframe to run analytics on competitive systems. So, the question to ask is, at what cost?
Our Eagle team
, which conducts free total cost of ownership studies worldwide, has found customers use 16 to 30 percent of their MIPS to move data off z Systems. We recently completed an Extract, Transform and Load (ETL) study quantifying the net cost of moving 1 TB of data daily from the z13 platform to an operational data store and then onto three data marts, all running on competitive analytical systems on distributed hardware. The study shows more than 88 percent ETL cost savings (more than $9M), keeping data in the zEnterprise Analytics System solution versus moving it out to the competitive systems.
Our sellers have access to an app to do a quick cost of ETL calculation for your environment and then follow up with a full customized Eagle study, if interested. To learn more about IBM’s winning solution for Analytics on z Systems platform, please read the point-of-view IBM Redbooks, "Reducing Data Movements Costs in z Systems Environments"
or my video talking about saving money by co-locating analytics with OLTP data
Since mobile adoption has gone viral, the best way to reach your customer with the best offer at the right time and right place is through mobile analytics. A recent article in IBM Systems Magazine, Mainframe edition discusses why the z Systems environment is an unparalleled home for mobile analytics.
Avijit Chatterjee, Ph.D., is a Mainframe Evangelist with the IBM Competitive Project Office.
» More Articles
The effects of a mass exodus from the mainframe ranks could affect a number of computing trends, not the least of which is big data.
Read more »
Mainframers give advice and lessons learned on what helps in the real world.
Read more »