Gain Analytics Agility on Apache Spark for IBM z Systems
When people discuss real-time big data domains, rarely do they understand the realities of where daily transactional data occurs around the world.
To give you an idea of the volumes we’re looking at, consider that for every second of the day worldwide, there are:
- 250 iTunes downloads
- 3600 Instagram photos taken
- 7000 Tweets sent
- 30000 Facebook likes
- 46000 YouTube viewings
- 60000 Google searches
- And 1.15 million CICS transactions on System z
System z is best positioned to handle the predicted explosion in social and mobile traffic from new and varied devices.
IBM z Systems customers that want to engage in value-add Big Data & Analytics initiatives can now do so with ease. The secret? They can run their analytics right where the preponderance of their transactional data resides right now. This is typically on an IBM mainframe since up to 80% of all transactional data interacts with z Systems. In the past all companies interested in extracting value from analytics with any systems platform had to engage with the cumbersome Extract-Transform-and Load (ETL) processes before gaining insights. Today those steps are no longer necessary. Companies can leave all the data right where it is now and save time, money, and headache. In fact a recent study shows that clients can save up to $10M in costs over 4 years just by eliminating ETL movement. So what makes all this possible? IBM z/OS Platform for Apache Spark released in March of 2016 offers enterprise-grade Apache Spark running natively on z/OS and provides data abstraction and virtualization to enable efficient access to a broad set of structured and unstructured data sources through Spark APIs. This allows analyzing data in place, securely and at scale, while avoiding costly data movement between servers.
Please do not take our word for it – watch this three minute video to hear why the marketplace considers Apache Spark on z as a game changer in the world of advanced analytics.
To learn more visit our IBM solution page for more details.
You may also be interested in the IBM Cloud Platform story around z Systems.