Servers

New release of IBM's z/OS tightens the lid on mainframe security


IBM z/OSIt appears that IBM has just introduced a new release of its renowned z/OS mainframe operating system.

Because of the mainframe's place in the heart of a vast portion of the world's financial services - as well as varied other large businesses, the focus this time round has been on security.

Excerpt from the IBM press release:

... the new version of z/OS to deliver improved network security and policy management, enhanced PKI (public-key infrastructure) services and adoption of the FKCS (Public-Key Cryptography Standards) #11 standard. IBM also announced new mainframe software that automates security adminsitration and audit processes.

Jim Porell, IBM Distinguished Engineer and chief architect of the System z, was quoted in eWeek:

Our security leadership is one of the many reasons why the world's top banks rely on the IBM mainframe for their financial transactions.

Additionally, enhancements were made to the ability of z/OS to deliver more robust scalability and availability for clustered environments. The IBM mainframe system can support up to 54 engines in a single z/OS image, scaling up to 1,728 mainframe engines behaving as one single system via a Parallel Sysplex cluster consisting of 32 systems.

Adding credence to what the company has been calling the renaissance of the mainframe, it is noted that the company's second-quarter of 2007 saw the fifth consecutive growth for the mainframe. This has only happened once since the early 1990s.

Do you think that the mainframe will ever die out?

--------------------------------------------------------------------------------

Stay on top of the latest tech news

Get this news story and many more by subscribing to our free IT News Digest newsletter, delivered each weekday. Automatically sign up today!

About

Paul Mah is a writer and blogger who lives in Singapore, where he has worked for a number of years in various capacities within the IT industry. Paul enjoys tinkering with tech gadgets, smartphones, and networking devices.

16 comments
Shoobe
Shoobe

I have supported IBM Mainframe Operating Systems for over 20 yrs. so I am a bigot for big iron. The IBM mainframe becomes more robust with each new model and each release of the operating system from OS/VS to MVT to MVS to OS/390 to z/OS. Stewart Alsop's foolish prediction in 1991 that all mainframes would be unplugged by 1995 was indeed foolish and hopelessly uninformed. There were many companies that drank the client-server-koolaide and attempted to move off of mainframes and indeed, there are several companies that no longer manufacture mainframes. After 10 years of trying to 'make it happen' many of those companies realized they weren't even close to anticipated ROI for all that work and the stablility required for business simply wasn't there. So back to the mainframe they went. They realized time lost by 're-booting' interruptions wasn't such a good thing. If we do our jobs correctly, there is no need to 're-boot' or IPL as we prefer to call it (Initial Program Load). The most troubling part is the greying of the support staff - We used to be called dinosaurs - now we are called sages. Even IBM has a sense of humor. When the IBM z990 mainframe was announced it was dubbed, by IBM, as the T-Rex. The younger generations need to start learning to replace us. The operating system has in someways become 'a point and click' configuration (as insulting as that is). I would be more concerned about the lack of Mainframe Developers than I would about the platform itselt. I have witnessed the decline of the American Programmer - and our expectations of them.

balbino_ph
balbino_ph

IBM mainframe will never die, it will just stay old.

paulmah
paulmah

Do you think that the mainframe will ever die out?

srajput123
srajput123

The worldwide installed base of IBM mainframe exceeded 14 million MIPS* by the end of 2008 ? up approximately 20% from year-end 2007 and roughly double of what it was five years ago. In the mainframe world, MIPS growth is a good indicator of actual growth because users pay for systems based on MIPS capacity and the MIPS growth is based on incremental MIPS added. It is clear that mission critical systems will continue to run on Mainframes & organizations will not risk moving these applications to alternative plat-forms like UNIX or .Net. Considering the hardware cost per MIPS is around $1500 to $3000 per year, for an average MIPS capacity of 2000 MIPS, the hardware MIPS cost alone will come to a whopping $3M to $6M Dollars annu-ally! A saving of 100 MIPS per year will translate to a saving of $150,000 K to $300,000 K in dollar terms. In light of the above analysis and the single point agenda of ?cost contain-ment‟, MIPS reduction becomes a viable strategy to reduce the Total Cost of Operation (TCO).

bogdan_pilawski
bogdan_pilawski

No one will attempt to deliver a vast quantity of goods across a continent in a small van, and no one will use a huge lorry to bring a daily pack of newspapers to a newsagent shop/kiosk/stand. There is and there always will be a place for a PC and for a mainframe.

dlandry
dlandry

As long as mainframes continue to make sense they will be used. The "main node" comes into play in many areas, distributed computing, clusters, networking... I think they will always have their place somewhere, same where P2P, etc has their places and will not die. Things will evolve, but I think certain fundamental computer science elements like main nodes will not.

Andy Goss
Andy Goss

With Linux and Solaris on IBM mainframes (does anyone else still make them?), the distinction, from the user viewpoint, will be academic. The mainframe always was just a big computer anyway, but with a rather horrible OS to keep the end users out of harm's way. I would expect that for situations where really huge databases are updated and accessed in real time, really huge, fast, monolithic computing will be inevitable, but I'm not about to place bets on what it will look like on the floor - one big box or lots of little ones.

KlausJW
KlausJW

As long as mainframes are able to deliver a value to business, they will not die out. They will always have a place in corporate IT, at least in a multi-tier architecture.

wrlang
wrlang

I've worked on all platforms and each has their advantages. Some midrange platforms have reached the reliability of the mainframe. Most smaller platforms continue to struggle to get into the 1980s with regards to reliability. Its hard not to bust out laughing when I see a fresh face all agog over the brand new concept of virtual machines. Ever hear of CP67? LOL Memory leak? Ever hear of a protection exception? I've been dreaming of GRIDs for sometime, had a few going back in the 80s. It won't kill the mainframe, but will change the complexion of computing.

algerg
algerg

The IBM Mainframe has been the staple of computing for as long as I have been working on them (my first was a IBM 360-40) back in 1977 to my last IBM 390. The computing world has gone full circle from mainframes, to servers, to local systems, now moving back to mainframes. The security and stability of mainframes are still the best and the costs of just having a thin-net terminal on your desk is effective and secure.

merichse
merichse

Midrange systems and servers are working hard to achieve the same levels of resilience, which you can get from the closely coupled interaction between hardware, software, and culture on this platform. Over time they will probably converge. But that might easily take another ten years. Partitioning, virtualization and so on are mainframe technologies being adapted by the other platforms - and SOA and Web are being adopted the other way round. I think that the difference in culture between those growing up with the knowledge of thousands of users depending on the stability of the system and those growing up with a single-user machine, where Ctl-Alt-Del did not disturb anybody else. That seems to be very hard to adopt outside the mainframe world for the moment, but eventually that will also need to converge. Otherwise we will not be able to develop our economies and societies like we want to. Best rgds, Michael (Who spends half of his time in Jurassic Park and the other half at the bleeding edge of technologies and architectures)

lmarks
lmarks

The IBM mainframe OSs (Z-OS and Z-VM) are actually quite elegant. One shouldn't misjudge them because of the rudimentary system utilities (text editors, mail agents, file-send agents). Here's a quote from the 1989 edition of Silberschatz & Peterson's "Operating Systems Concepts." "The view of the operating system seen by most users is thus defined by its systems programs, not by its actual system calls. Consequently this view may be quite removed from the actual system. The problems of designing a useful and friendly user interface are many, but they are not direct functions of the operating system."

Shoobe
Shoobe

MVS....Multiple Virtual System - and they think it new! Dual Processors? woo hoo they got 2! Try the new z9-EC 54-way.

macumazahn
macumazahn

Worked on mainframes from 1974-1999, we were moving forwards with mainframe and a little client-server apps, after phasing out mainframes to save money, as much or more was spent on pcs maintenance, equipment, upkeep, service than on the mainframes, and they are less reliable, and more prone to upgrade after upgrade, give me back a mainframe

bstockha
bstockha

The main-frame's ability to truly multi-task large, long-running jobs without the machine/OS "hanging", and it's ability to quickly move huge amounts of data to and from channel-attached storage devices will ensure its staying power for years to come.