In this series, Janet Sun explores and dispels numerous myths around the mainframe. Continued from Part 1.
Mainframes are Expensive?
Think that mainframes are expensive? That depends on what you’re looking at.
If you are looking at acquisition hardware costs, certainly a single mainframe costs more than a single server or even several servers. But, you would certainly need more individual servers to match the compute capability of a mainframe. Add to that the idea that software and labor costs for servers grows linearly – the more servers you add, the more software licenses and the more system administrators are needed. And yet, the mainframe delivers higher utilization, lower overheads, and the lowest total cost-per-user of any platform. When all cost factors are considered fairly, the mainframe is usually the lowest cost alternative.
Often when considering the cost of the mainframe, we only look at the initial hardware purchase and overlook the on-going maintenance costs. If you have a hundred servers, that’s a hundred times more chances that something is going to break, so you need an army that has to be ready at any time to fix hardware.
Each of those servers has an operating system on it, and all of them need patches, upgrades, and applications deployed to them on a regular basis. So, you need another army doing all of that.
Then your applications are spread all over the place, so when the software fails or gets overloaded, it takes an army of people to monitor the applications and figure out where the problem is. Servers are cheap to buy, but then your savings get totally blown away by needing all of these people to run and monitor them.
And don’t forget electrical and air conditioning costs also increase when you add servers. Then, you need to make sure that you count ALL the servers?
There is a story of one company that thought that it had only 24 UNIX servers. On further examination, it turned out those were just the production servers. Adding development, test, and failover servers more than doubled the count yielding 49 servers. So, the hardware, software, energy, and labor costs were higher than originally calculated.
The Skills to Manage Mainframes are not Available or You Need More People
As we have already seen, it takes fewer people to manage a mainframe than it does a set of servers delivering comparable capability. Do you need specialized skills to manage a mainframe? It depends.
If you are managing Linux on the mainframe, you will find that Linux is Linux regardless of platform, so if you can manage Linux on Intel, you can manage Linux on the mainframe. This means those students coming out of universities that know Linux can, with very little additional training, manage a Linux on the mainframe environment.
In addition, IBM has been investing in increasing the available skills. The IBM System z Academic Initiative ensures that a System z and z/OS skills shortage does not happen. Since 2004, the program has worked with 1,000 schools to educate more than 50,000 students worldwide! Many people in the mainframe community are using the System z Academic Initiative to assist and enable schools to teach mainframe skills.
Then there is SHARE’s zNextGen community that connects more than 900 young mainframe professionals from over 24 countries. And don’t forget the two annual SHARE conferences and year-round webcasts, which offer hundreds of hours of mainframe skills training and numerous opportunities for peer networking.
There is plenty of access to the skills required to manage a mainframe and scores of experts who are happy to share their knowledge and experience. Being well-versed in mainframe technologies is looking like a pretty good career choice for those looking to make a valuable contribution to any one of a number of industries or even society in general.
Why Attack the Mainframe?
In the early 90’s pundits predicted the end of the mainframe, and the media piled on.
The mainframe was an easy target. It was not well understood by the general public (and still isn’t, as demonstrated by the video and HuffPost article); it has many detractors and few, if any, proponents.
IBM almost stands alone defending the mainframe, and most of its arguments are dismissed as marketing hype. As the mainframe community, we at SHARE support the value that the mainframe delivers and can substantiate IBM’s claims.
Are companies running from the mainframe? There are certainly a few who are contemplating or attempting to migrate off the mainframe. But the work being done by mainframes is increasing, not decreasing. Just looking IBM’s Annual Reports – 2010 shows a 22 percent increase in MIPS (capacity) shipped over the previous year, 2011 16 percent increase, 2012 19 percent, and continued growth every year, including this year.
Migrating Off the Mainframe?
Moving off the mainframe is not an easy task and should not be taken lightly. It is likely a multi-year, multi-million dollar effort that is fraught with risk.
You’ve probably seen headlines where a CIO has announced plans to move off the mainframe. You’re less likely to see a headline where a CIO admits that they made a mistake and are cancelling a project to move off the mainframe.
There’s the example of the company which budgeted $10 million for a one-year migration from a mainframe to a distributed environment. Eighteen months into the project, already six months more than planned, the company had spent $25 million and only managed to offload 10 percent of the workload. In addition, it had to increase staff to cover the over-run, implement steps to replace mainframe automation, acquire additional distributed capacity over the initial prediction (even though only 10 percent had been moved so far), and extend the dual-running period (at even more cost due to the schedule overrun). Not surprisingly, the executive sponsor is no longer there.
What is the business purpose for moving? Herd mentality? Everyone else is doing it? We’ve already seen that many companies still use mainframes, and are continuing to invest in mainframes.
Lower costs? Which costs? We’ve seen that looking at Total Cost of Ownership (TCO), including on-going software licenses, maintenance, energy and air conditioning costs, and labor costs, the mainframe is often less expensive. IBM has conducted nearly 100 studies comparing costs for companies that have considered re-hosting applications on distributed servers. The average cost of the distributed alternative is 2.2 times more than the mainframe. Only 4 cases showed lower costs for distributed servers.
What about switching costs? New hardware, new software, new processes, testing, training. Not cheap. What was the reason to migrate off? What could possibly go wrong? You can’t replicate the same capabilities of the mainframe versions of the applications on the distributed platform? Re-coded applications don’t quite work? Other unanticipated problems? What about the other mainframe attributes that you give up?
Ready for more mainframe misconceptions? Janet Sun continues the Don’t Believe the Myths about the Mainframe series with the forthcoming Part 3, very soon on Planet Mainframe.
Originally published on the SHARE.org blog.
Janet Sun is Senior Consultant, at Sun Coast, responsible for z/OS storage and cloud portfolios.