One of the best parts of the job is the opportunity to regularly meet with IT organizations and talk shop. For a geek who’s been doing this for a while it’s super important to stay connected to users and get a real world prospective of what’s going on in their environments and where they are headed.
No question, the trend to modernize infrastructure is top of mind. Discussions include strategies around deploying cool technologies like; Flash for all primary storage. Scale out systems that can start small and get big. Converged platforms to speed deployments. And pretty much software defined everything to support a growing subset of workloads. While a lot of this technology is applied to new apps running in cool and hip virtualized environments, there’s still a very healthy demand to apply those capabilities to those super important traditional apps running in physical environments. This includes major efforts to modernize the Mainframe.
Fun Facts – Did You Know?
Yup, Mainframe is still the heart of financial and transactional systems. Whether you realize it or not, we all rely on it every day. It’s responsible 23 Billion (with a B) ATM transactions a year. In fact, there are 60X more transactions per second completed on a mainframe compared to all internet transactions. Just in credit cards, $6 Trillion (with a T) is cleared per year, which is totally believable based on inspection of the Delandy household’s monthly Amex bill.
But did you know supporting Mainframe from a storage perspective requires a fairly significant investment in R&D and Mainframe skills? This is a big reason why there’s also a fairly small club of providers. Dell EMC has invested $ Billions in Mainframe for almost 3 decades, and those investments continue. And during that time there is a long list of Dell EMC innovation and “Firsts”. Many of these capabilities are now also applied to open and virtualized environments as well. It includes capabilities such as replication, snapshots, and fully automated multi-site failover and failback. All capabilities that help run the world’s most mission critical apps.
When people think Mainframe, they think large organizations such as financial services, government agencies, and airlines who run the world’s most mission critical applications. These are the folks who have invested heavily in their Mainframe environments and will continue to do so indefinitely. They are also a relatively exclusive club of a small number of organizations, but the volume of transactions and amount of data being processed is huge. The need for massive scale not only includes capacity and performance, but better ways to manage those huge datasets.
A good example of the issue of scale is from one of our Mainframe users runs “zDP”. It’s a VMAX snapshot technology that lets them create lots of copies, so if data get corrupted or deleted they can recover it near instantly. Think of a DVR for your storage array where you can rewind to any point and hit play. What’s cool is its pointer based, and not a full copy, so its mind blowingly space efficient. In this case zDP to protect 23,000 volumes with snaps every 10 minutes to meet their recovery point objectives as part of their service level. When you do the math, that’s 3.3M snaps per day. It’s a lot of data. So size does matter.
But you might be surprised to know that there are lots of smaller organizations such as retailers, insurance agencies, and transportation companies that also rely on the Mainframe as well. The data sets may not be as large, relative to other parts of their environment, but the apps are absolutely mission critical to their business. For example, I recently met with a smaller insurance company that deals exclusively with independent brokers. The app that processes payment to the brokers is relatively small, about 10TB. But if the app is not available, their brokers don’t get paid. And because the app is highly integrated into their operations and work flow, moving to something else would be extremely costly and risky. So in a way, size sometimes doesn’t matter.
What’s New and Cool?
Many of our user are looking at ways to modernize their Mainframes to provide more efficiency and better ROI on their investments in these apps. The new VMAX 950 has many users excited about some of the really cool Mainframe capabilities. Here’s the short list of the key areas of coolness.
First is more performance and better scale. The VMAX 950 is an all flash array, which is optimal for Mainframe apps where low latency for transactions is king. In addition, the new Z-bricks offer more CPU horse power and faster cores to driver even higher I/O rates. Under the covers, it provides better optimization for FICON connectivity which in many cases provide lots more performance with lots less hardware. And the new packaging provides an even smaller and more economical starting point of 11TB, which just so happens to be the perfect amount of capacity for storing your precious family photos on your home mainframe.
Second, is great for consolidation. Providing more performance and capacity in a much smaller system can significantly reduce foot print and lower TCO vs older arrays. In fact just the power and space saving costs compared to a 3 or 4 year old array can pay for the new VMAX system in itself. There’s also the ability to intermix environments by combining open systems and mainframes into a single storage system. Which is an awesome capability for users with a small amount of mainframe data, because consolidating it with the open side is way more costs cost effective than to keep the mainframe storage as standalone.
And last, but not least, its super simple. Lots more automation, single flash tier to manage, a single replication technology for mainframe and open, all good stuff. But it’s also simple for IT to consume. The new zF and zFX software suites make software licensing easy and way more cost effective. As we learned when we first introduced the F and FX packages, users love the simple packaging, now for the MF too, includes zDP THE killer app for data protection and recovery. From a Dell EMC perspective, no other vendor provide a similar capability.
Yes, the Mainframe continues to live on as a key part of infrastructure for some organizations. The types of apps running in this systems are perfect match for the performance, scale, reliability and trust delivered by VMAX. And it an area we continue to make significant strategic investments to deliver innovation for our users now and in the future.