The Financial And Environmental Benefits Of Using Mainframes


There comes a time for every organisation when they need to examine where they are and where they want to be in the future. When they look at what they are doing well and what their competitors seem to be doing better. Quite often this involves senior management going away for the weekend together, and quite often disagreements can be based more on beliefs rather than the facts.

One area that often receives a lot of attention, but suffers from the heat of almost religious beliefs about what’s right and what’s wrong, is IT. You have the Linux enthusiasts, those who believe technology should begin with the letter “i”, Windows zealots, and people sometimes treated as being out-of-date who recommend mainframes!

But let’s examine this rationally in the cold light of day. Logically, there are really only two reasons to change technology – it’s better or it’s cheaper. And now there’s a third – it’s greener.

Saving the planet, reducing CO2 emissions, and cutting energy demands all work for the green agenda – something that is gradually making organisations more attractive to other businesses as well as the person in the street – can also come with a huge financial benefit.

There are many sites that have allowed their IT to grow organically over the years, meaning that they currently have a number of different platforms and software, each ideally suited to the computing requirements of different periods in the organisation’s past.

These servers manage about 20 per cent utilisation. More recently, a degree of control has been introduced with the introduction of virtualised servers, but even so, many sites are experiencing virtual server sprawl.

A mainframe can run at 100 per cent utilisation (and usually runs at somewhere between 80 and 100 per cent). And many organisations new to virtualisation are amazed to find that mainframe virtualisation has been available since the 1960s.

This removal of server sprawl means that less air conditioning is required than to cool the sprawl. Without all those servers, there is much less heat being produced, so much less coolant is required in the air conditioning. And this will reduce CO2 emissions, which are said to be causing increases in average global surface temperatures, a rise in sea levels, and a global change in precipitation patterns.

The other obvious benefit in reducing server sprawl is the cost savings, because less energy is needed to power them. There’s also a saving in the amount of space needed for IT. This means that less building space needs to be rented or the space where the servers were kept can re-used by other departments that would have needed to rent extra office space – leading to further cost savings.

Older Linux, Unix, and potentially Windows servers would have a cost associated with them for necessary maintenance and possibly replacement. Migrating to a mainframe saves on the cost of the many different experienced engineers potentially needed to maintain these different servers.

There’s also a cost saving in down time. These old servers would be unavailable during maintenance so no users could be working on them. On a mainframe, there would be a single engineer required, less likelihood of maintenance being needed, and, when it was, the service would probably still be available to users.

It might be difficult to be able to scale up an application running on a particular distributed server. Migrating to a mainframe makes scalability no longer an issue. This again leads to cost savings. And in terms of failures and required maintenance, mainframes are known for reliability and 99.999 per cent up-time. Again, massive savings to the bottom line for an organisation.

Migrating Linux systems onto a mainframe is something that many organisations are reporting as being so successful they wish they’d done it years ago. Mainframe Linux has been available for just over ten years, but is only now being recognised for the benefits it offers.

Linux on System z runs as a native operating system, there’s no emulation involved, although many sites run zLinux under z/VM, which allows any number of virtualised Linux systems to be available. One of the big advantages of running Linux on a mainframe is the IFL (Integrated Facility for Linux) specialty processor, which can run Linux applications like a central processor, but without the usage being charged for – which is what happens on central processors.

There is a charge for the IFL specialty processor. With this pricing model, an organisation can increase the amount of work processed on its IFL without incurring any additional charges – again resulting in financial savings.

IBM’s new z114 hybrid system can now run Windows on its x86 blades. This means that not only Unix servers, but Windows servers can be removed from around the organisation and everything can be run and controlled from a single hybrid box. The savings on space, air conditioning, power, maintenance, and replacement now apply pretty much across the board.

Thinking the unthinkable for some people means that an organisation’s balance sheet will show savings with the migration to a mainframe now. Plus there will be environmental benefits for the migration – meaning a company can promote its green credentials and potentially win more business because of them.

Marcel den Hartog is Principal Product Manager EMEA for CA’s Mainframe solutions. In this role, he is a frequent speaker on both internal (customer) and external events where he talks about CA’s mainframe strategy, vision and market trends. Marcel joined CA in 1986 as a Pre-sales consultant. Before this, he worked as a programmer/systems analyst on VSE and MVS systems, starting with CICS DL1/IMS and later with DB2. He is still an expert in CA Easytrieve and Cobol and has hands-on experience with many CA products. He was responsible for managing CA’s pre-sales teams in The Netherlands, Belgium and South Africa for a number of years. Prior to his current role Marcel worked as a Linux Development Architect for CA’s Linux and Open Source team. In that role, he served almost two years as a board member of the Plone Open Source Community.