You don’t need to throw the baby out with the bathwater... when you can modernize with IBM Power Systems!
During my career, I have been fortunate enough to work with some really cool tech. From early on in my career at Newcastle University working on SuperJanet and Cray, through to my time at FIS building supercomputers for actuarial modeling as SaaS and deployed directly to corporate clients across the globe.
'Cool tech' - or rather technology that can deliver significant business advantage - has always started my head buzzing, thinking of use cases to benefit the organizations I work with.
Whilst an avid technologist, I have learned to temper my enthusiasm at first, as I've been around technology long enough to be a realist about the economic benefits (some would say skeptic at times!), but I prefer to think of myself as a 'technology pragmatist' nowadays.
In my experience, most technologies that have made the successful transition from conceptual ideas to commercial products have some underlying usefulness. It all comes down to choosing and backing the right horse for the right course, and I pride myself on being open-minded when it comes to technology, and selecting horses!
Although I consider myself to have had a pretty varied career, working with Windows, Linux, and AIX through the last 2 decades, I have to confess, my experience with IBM Power Systems (before joining Tectrade/CSI Group) was limited to a 3-year stint with an American corporation over 20 years ago!
I didn’t work directly on the AS400, but as Wintel lead back in the late ’90s, my 'Pod' was directly next to the AS400 tech lead's 'Pod'. Every day I would hear terms like LPAR’s, Batch Jobs, and Robotic Libraries.
Occasionally an IBM engineer would appear to change a part that had failed or was about to fail - even then, the AS400 dialed back home to report faults!
One thing that was clear though, the AS400 was the crown jewel in that organization’s IT estate.
Rock-solid uptime and performance to handle anything thrown at it were the order of the day.
In this instance, the AS400 ran a finance package called Movex, which was similar to SAP at the time. Overnight it would run batch jobs and provide basic reporting.
It was reliable and handled the accounting, production control, and supply chain for 120 business entities in Europe.
However, as the business' appetite for better forecasting information increased it became apparent that Movex alone could not provide what was required. The answer was to reinforce the Movex reporting capabilities with Cognos running on x86 systems and Windows.
During this epic project, the business was structured into 4 divisions; (each containing approximately 30 separate companies), and as you can imagine, the resulting project was expensive due to its scale and performance requirements.
Strict rules around performance, size of cubes, and data extraction meant the final design called for a 6 node SQL server enterprise cluster farm and supporting 32 core Cognos Servers, one dedicated to each division.
Figure 1.0 Cognos Reporting High-Level Logical Architecture
The solution in the overview diagram above had the following characteristics;
You've heard the saying 'can't teach an old dog new tricks' so why is this “Old Dog” blogging about an AS400 design pattern from 2 decades ago?
Well as you probably know the basic concepts of IT are like clothing. If you wait long enough, it will come back into fashion!
We saw this with centralized computing, then distributed systems, and back to centralized with public cloud, VDI, etc. With edge computing, traditional on-premises computing, and public cloud - we have the hybrid world we live in (and realistically will be for some time).
I recently co-hosted a presentation with Paul Chapman from IBM & Red Hat, who works alongside Tectrade to support our IBM clients as they modernize their environments.
The presentation involved a demo of a core banking application running on an LPAR consisting of a DB2 database and a front-end application (see below video). Although the application has rock-solid uptime, its interface and functionality were pretty basic and didn’t allow the business to develop in the ways it wanted to.
What should the business do? They are happy with platform performance, uptime, and scalability, the issue is that it isn’t agile enough… or so you would think!
So here's the crux... you don't need to throw the baby out with the bathwater. Application modernization allows you to reinforce your solution by adding new functionality to an existing environment without affecting the core codebase.
In a typical application modernization approach, you have the 5 R’s - namely:
If you don’t own the codebase or cannot change it, you might think that the only option left is to "Replace", but there is actually a 6th option to "Reinforce".
Rather than the extremely difficult task of removing your core system, you can supplement it with additional functionality using exactly the same methodology from 20 years ago.
The methodology has been modernized through the IBM and Red Hat relationship using the latest technology stack of containers and OpenShift running directly on an IBM Power Systems LPAR alongside the core banking app and DB2 database.
There is now no requirement for expensive clusters, VM’s and SAN’s either. Plus, with our private Tectrade PowerCloud hosted in Tierpoint data centers, we can attach your legacy workloads, seamlessly and securely to IBM, Microsoft Azure, Amazon Web Services, or Google Cloud Platform, as and when you need them.
Don't take our word for it... we won the 2021 IBM Beacon Award for our hybrid cloud solution, which involved IBM Power Systems in the IBM Cloud.
If you would like to get in touch with Jason or any member of the team, please complete the contact form here.
Blog by Jason Normanton, Head of Cloud Services