On April 10, 2018, IBM announced their latest mainframe offering, the IBM z14 Model ZR1. This announcement came hot on the heels of the original z14, which was delivered less than a year earlier in July 2017. So why the new model? And what does it mean to mainframe users out there?
It is Smaller…
Yes, the big news about the ZR1 is its form factor. By that, of course, I mean its size. Analysts have dubbed the ZR1 the mini-mainframe and its smaller size is a compelling feature. You can purchase a rack-mounted z14 ZR1, which means it fits into a standard 19-inch rack that is compatible with the rest of your data center. For years, mainframe shops have accommodated the mainframe separately from the rest of their computing infrastructure. And it had to be that way because mainframes always required more space than their other servers.
So, the smaller size of the z14 ZR1 means you can start thinking about combining things and reducing the amount of space your mainframe requires. With the 16U of free space in the standard ZR1 configuration, you can add other servers, storage, networking and monitoring systems right in the same rack instead of in additional cabinets. For example, you can include zHyperLink capability right in the rack with the z14 ZR1.
By reducing floor space and combining mainframe and open systems servers in the same rack, you can reduce energy usage, too. The ZR1 uses standard air cooling and single-phase power, instead of the three-phase power supply of the original z14.
…but it is Still a z14
Size notwithstanding though, you need to understand that the IBM Model ZR1 model is still a z14. The same things that make the z14 noteworthy apply to this new model, too. That means you get pervasive encryption for true end-to-end protection of all your mainframe data, without interrupting business applications and operations. So what does that really mean for you?
Well, the first big problem when looking to encrypt data is figuring out which data should be encrypted. Think about that for a minute. Do you have metadata and documentation about the data in all of your mainframe data sets, databases, and files? Probably not. So how can you match up regulatory requirements to your actual data? Well, pervasive encryption solves that problem by efficiently encrypting all of your mainframe data. That can save a lot of time and effort in this day and age of increasing regulatory compliance requirements (e.g. GDPR).
But pervasive encryption is just one key feature of the z14. All of the value propositions of the z14 are just as relevant for the new Model ZR1. The speed and reliability you have come to expect from mainframe computing are built into the z14 ZR1. For example, consider speed and scalability features like IFCON Express16S+ which improves I/O rates, new SIMD instructions that boost all workloads, and the IBM zHyperLink Express, which uses a direct-connect short-distance link to deliver low-latency connectivity between the z14 and FICON storage systems.
Being a z14, the Model ZR1 also enables you to double the amount of real memory allowing up to 8 terabytes. Many organizations are interested in adding more memory to facilitate new and changing mainframe workloads due to increased mobile transactions, analytics processing and machine learning.
It is important to know that the rack-mounted z14 ZR1 is not the same as a standard configuration z14. It can have a maximum of 30 processor cores, instead of the 170 cores available in the standard configuration models. Of course, that means that the price of the ZR1 will be lower than a fully-configured standard z14.
Even so, the z14 ZR1 can deliver a significant amount of processing power with both traditional mainframe CPs as well as Linux IFLs. Organizations considering consolidating workloads can run more workload in a single frame on a z14 ZR1, thereby helping to reduce costs… all while staying on the mainframe platform.
In addition to the ZR1, which can run either z/OS or Linux, IBM also offers a version dedicated to Linux, named Rockhopper II. This means that mainframe users get to pick and choose the hardware, footprint, and operating systems that work best for their needs and implementations.
Cloudy with a Chance of Mainframe
The z14 Model ZR1 can be the base for a securable, integrated cloud. It offers many features that help to enable cloud computing, including Central Processor Assist for Cryptographic Functions (CPACF), Crypto Express6S for FIPS-4 certified encryption key management, Trusted Key Entry (TKE) systems, and enhanced applications of Secure Service Containers including for client applications.
The ability to deploy software using containers is fast becoming an industry standard, and with the ZR1 IBM is improving its logical partitioning capabilities with Secure Service Container (SSC) technology. The IBM SSC provides a cohesive, integrated platform with IBM Cloud Private for hybrid and private cloud deployments on z Systems and IBM LinuxONE servers.
But what does this mean to you? Organizations can deploy containerized IBM middleware applications as well as use common tooling for deploying homegrown or other third-party Docker and Kubernetes based applications. With SSC improvements for security, application management, and usability, combined with pervasive encryption the z14 ZR1 can deliver a protected, secure environment for running application workloads. Indeed, SSCs bring the mainframe squarely into the age of containerization – and that is a good thing for mainframe users.
Containers and Mainframe Pricing
Another key factor to consider is the ability to containerize mainframe workload and take advantage of IBM’s reduced container pricing plans. Container pricing helps to lower the cost of mainframe computing by treating development and testing separately, paying less for brand new mainframe workloads, and treating payment systems differently.
Organizations that rely on mainframes are always looking for ways to reduce costs and the new pricing options provide that. Additionally, container pricing may be able to forestall the movement of workload off of the mainframe. Instead of looking for ways to migrate workload off of the mainframe to reduce costs, you can containerize specific types of mainframe workload – and save on costs, without moving off the mainframe.
Nevertheless, mainframe pricing is still complex and difficult to understand. And adding three additional pricing methods doesn’t make things any simpler.
More to Consider
There is a lot to like about the new z14 ZR1. And that is good news for mainframe aficionados because the platform needs to be constantly rejuvenated and improved to make it compatible with modern computing needs. But an argument can be made that IBM needs to add more focus on simplifying mainframe usage and management.
Although many programmers interface with the mainframe using an IDE, performing ancillary tasks like cataloging a data set or checking on a job status require switching to a cumbersome ISPF interface. Anything that can simplify and modernize interfacing with the mainframe should be a high priority for IBM – and it should be included at no cost, as opposed to an add-on offering.
IBM is also spending a lot of effort on machine learning and AI, even offering the IBM Machine Learning for z/OS solution on the platform. That is great, but the mainframe is perhaps the most instrumented computing platform in the world, so there are a lot of metrics that could be automatically analyzed and used to improve the performance and management of the mainframe environment.
The Bottom Line
So, what can we take away from all of this? Well, IBM is obviously committed to innovating with its z Systems hardware, and that is good news for mainframe users. And the new small form factor of the z14 ZR1 can provide advantages to mainframe data centers with a mix of size and performance that can optimize performance, convenience, cost, and energy consumption.
Larry Strickland, Ph.D. is the Chief Products Officer of DataKinetics. He has been a C-level executive for years, making technology simple for the end user, and understanding the impact of technology on the user. He is passionate about product innovation that helps clients solve their most challenging issues while delivering solutions that transform their business.
Latest posts by Larry Strickland (see all)
- Analytics and Machine Learning in the Modern Datacenter - Oct 31, 2018
- IBM z14 Model ZR1: A Smaller Mainframe for a Larger World - Jun 26, 2018
- Open Source on Steroids - Nov 2, 2017