"The sheer transaction muscle of the new systems is impressive but also answers the order of magnitude increases in transactions that banks, financial institutions and credit card companies expect as individual sales trigger 'star bursts' of associated interactions," King said. "In essence, IBM has, yet again, re-imagined and reinvented the mainframe for a new generation of transaction processing. That should keep the platform relevant for IBM's dedicated enterprise mainframe customers and could also inspire other businesses to consider z13 solutions."
Moreover, IBM claims the z13 will feature the world's fastest microprocessor, twice as fast as the most common server processors, with 300 percent more memory and 100 percent more bandwidth and vector processing analytics to speed mobile transactions. IBM has designed the z13 to integrate real-time scoring and guarantees this capability as a feature of the system. This scoring can be used for fraud detection.
In addition to assistance with fraud prevention, businesses looking to enhance their customer loyalty programs will be able to use new z13 capabilities to add more personalization by gaining a real-time view of a client's purchasing habits to offer upsell and cross-sell promotions before they leave the store—and in some cases before they even enter. With the z13, businesses will be able to use IBM's predictive analytics modeling technology, SPSS, and personalize the transaction as it occurs.
"The big question in my mind now is how the market will react to these changes," said Joe Clabby, founder of Clabby Analytics. "The mainframe is now capable of delivering analytics results on transactional data in real time. And this means that data no longer needs to be ETLed to distributed processors. But the ETL [Extract, Transform, Load] process has been instantiated over a couple of decades—so people are used to moving their data to other processors for analysis. Plus, those same people are usually 'distributed systems' customers who will not be exactly thrilled to see work moving out of the distributed computing silo to mainframe environments."
The mainframe includes new support for Hadoop, enabling unstructured data to be analyzed in the system. Other analytics advances include faster acceleration of queries by adding DB2 BLU for Linux, providing an in-memory database, enhancements to the IBM DB2 analytics accelerator and vastly improved performance for mathematically intense analytics workloads, IBM said.
The z13's cloud architecture has the ability to scale and reliably and securely handle multiple workloads. In a scale-out model, it is capable of running up to 8,000 virtual servers—more than 50 virtual servers per core, allowing for lower software, energy and facilities costs, IBM said. In addition, IBM said the z13 lowers the cost of running cloud on the mainframe to almost half the cost of running cloud on an x86/distributed server environment, while boosting performance by up to 30 percent. Additionally, the z13 is an open platform, fully supporting Linux, OpenStack and more.
As part of the z13 announcement, IBM also previewed new z/OS software that delivers advanced analytic and data serving capabilities. This new operating system expands the ability of the z13 to process in-memory analytics and provide analysis on mobile transactions, helping clients further extend mainframe enterprise applications to the mobile user.
A new mainframe represents a significant boost to the IBM sales cycle, and the z13, with its focus on supporting mobile, cloud and analytics workloads, signifies IBM's push to shift more of its focus to higher growth markets. The move also indicates that the mainframe remains relevant in a mobile-first, cloud-first world. This launch complements IBM's ongoing investments to help clients drive mobile innovation across the enterprise.
"The reason that it makes sense to push to a mainframe architecture for tablet and smartphone loads is that these devices are increasingly connected to ever faster services and, to preserve battery and assure security, much of the data and processing power has to be in the cloud," said Rob Enderle, founder of the Enderle Group. "If you look at Seri and Cortana, these two differentiating services are hosted and don't work if the device isn't connected."