Being a consultant specializing on performance issues exposes me to a wide variety and types of systems and platforms. Focusing on performance issues on a variety of platforms, cloud, distributed, and mainframe, I’ve realized they all have wonderful uniqueness and plenty of issues to keep me mystified, curious, and entertained. It is always interesting to see another new generation of college graduates (mainly millennials) discover that basic computing principles continue to be paramount and are usually at the heart of all performance issues. The principals of CPU power, memory usage, and I/O continue to be ignored during the early phases of a project, and questionable design decisions are made on every platform, in all programming languages, and in new and old applications. This is why performance consulting continues to thrive even in these interesting technology times.
Given colleges’ course bias toward the new technology of cloud and distributed technologies, it is understandable why the new millennials and some of their managers don’t like the mainframe. Some managers and millennials have never really worked or experienced processing on a mainframe when they leave college. Some colleges don’t even teach mainframe level courses. They simply don’t know anything about a mainframe, its complex array of state of the art chip technologies, the myriad of system and application settings, the large governance audience, its sometime bureaucratic nature, and the shared responsibility of application goals. So below are a couple of ideas that have been somewhat successful in helping educate and help people new to the mainframe how to deal with agile development schedules, deployment, and support of their applications on the mainframe.
- It’s all the same it’s just not all about you and your application.
In the age of the selfie and the millennial “me” generation the cloud and distributed personalized environments continue to be encouraged because that is what the colleges have endorsed. Agile development people can be more focused when they can control, understand, and sometimes directly, or with a very small team, try to fix all the issues. The personalized distributed and cloud systems are sometimes managed by the agile development team and/or only a single system administrator so they easily understand where to go for help, advice, and fixing any issue. If the system administrator can’t help. he can find someone else that can. Performance is fixed by adding a server, memory, or storage; easily done with business user’s approval or a credit card.
All application lifecycle aspects, including resource provisioning, configuration management, application deployment, software updates, monitoring, and access control need to be done for all and any kind of systems and applications. Within the agile distributed and cloud environments the settings rarely impacts another business unit or application. In the shared resource of the mainframe all of those agile development processes have checks and approvals because of the shared mainframe environment.
It is very important to understand your complete and overall agile project and development user stories so that your agile project can go to the mainframe change control committee to schedule your implementation, configuration changes, and resources necessary for your agile project. Also, the agile team needs to meet all the other teams that are asking the mainframe support groups to do changes for them also. Meeting all the other teams and the large number of initiatives wanting the mainframe support people’s time and knowledge can be an eye opening experience for the small selfie agile team to realize that it is not all about their little project. - Every type of application or application development language can be done on a mainframe.
Social, mobile, analytics, any application anything you can think of runs on a mainframe today. From assembler COBOL, FORTRAN, JAVA, .NET, C++ and almost every other programing language can be run on the mainframe. It would be hard to find any programming language that could not be run on the current mainframe. Anything you can do on cloud or distributed systems you can do on the mainframe and sometimes it is easier and faster.
Many of the new phone mobile and social applications run on the mainframe. The new mainframe people need to understand the mainframe, the original computing platform, is just another server, cloud or sometimes even a distributed system. It runs everything and anything and deals with all the same issues the first cloud and distributed systems do. The mainframe just does it on a bigger scale, a larger number of applications and uses a larger number of settings to make sure each of its many applications get the best processing performance.
So given that the buzzword bingo says that everything needs to be “cloud first” why doesn’t your application go to the first cloud, the mainframe? The mainframe is the “everything” computing platform, so start experiencing its reliability, availability, and scalability today.
Dave Beulke is a system strategist, application architect, and performance expert specializing in Big Data, data warehouses, and high performance internet business solutions. He is an IBM Gold Consultant, Information Champion, President of DAMA-NCR, former President of International DB2 User Group, and frequent speaker at national and international conferences. His architectures, designs, and performance tuning techniques help organization better leverage their information assets, saving millions in processing costs.
Leave a Reply