Microsoft-Backed Firm Says IBM Is Anticompetitive 174
BBCWatcher writes "Microsoft has long claimed that the mainframe is dead, slain by the company's Windows monopoly. Yet, apparently without any mirror nearby, Microsoft is now complaining through the Microsoft-funded Computer & Communications Industry Association that not only are mainframes not dead, but IBM is so anticompetitive that governments should intervene in the hyper-competitive server market. The Wall Street Journal reports that Microsoft is worried that the trend toward cloud computing is introducing competition to the Windows franchise, favoring better-positioned companies including IBM and Cisco. HP now talks about almost nothing but the IBM mainframe, with no Tukwila CPUs to sell until 2010. The global recession is encouraging more mainframe adoption as businesses slash IT costs, dominated by labor costs, and improve business execution. In 2008, IBM mainframe revenues rose 12.5% even whilst mainframe prices fell. (IBM shipped 25% more mainframe capacity than in 2007. Other server sales reports are not so good.) IBM mainframes can run multiple operating systems concurrently, including Linux and, more recently, OpenSolaris."
Re:Buh buh but.... (Score:2, Informative)
The mainframe of old - the single room-size unit with hundreds of CPU's, drives and memory is indeed dead. These days a 'mainframe' is nothing more but a clustered Linux environment that runs virtualized instances of an Operating System. Some mainframes still resemble the old mainframes (eg. the zSeries) but they take up about the size of a rack.
Re:I Don't Quite Understand (Score:5, Informative)
1) Reliability: 5 9's (99.999%)
2) Backward compatibility, there are people still running applications written 40 years ago
3) Security: Physical (hard to move a refrigerator), Network (no external network when applications working internally), RACF, Highest level of security rating of ANY server, ever.
4) Architecture: Redundant everything: Spare processors, spare power, spare, everything. Predictive failure/automatic fail over for individual components. Memory Bus greater than anything out there. Pipes to Storage extreme. Cryptographic processors to do SSL, etc.
5) Scale up: 64 processors (4.4GHz), 1.5 TB of Memory, etc.
6) Scale out: GDPS (Geographically Disperse Parallel Sysplex) up to 32 boxes?
7) Hipervisor: Its a network in a box. Applications talking to each other use IP, not TCP/IP, so you aren't sending 35% data, 65% header when applications talk. Network is at the speed of memory. zVM has been developed for over 20 years.
8) Power Efficiency: Compared to a server cluster + cooling + redundant power, etc.
9) Network Simplicity: No need for a rats nest for your rack, cable simplicity in some cases from over 1000 cables down to 12. From 14 switches (which are very expensive) to 4.
10) Management simplicity: Less staff needed to keep it up and running. Instead they are focused on adding business value
11) Running Legacy (aka Business Critical) applications, your web presence, your portal, and a myriad of other disparate applications in one place.
12) Create new servers in minutes without needing hardware "on standby."
13) Compartmentalization in a single box
14) Shared everything while still fully separate
15) Workload manager: able to on the fly change how much resources are allocated to images AND (this is the cool thing, cause other VMs do that) give it goal times for operations. As in: Complete this task in 1/100th of a second, and it will allocate, on the fly, for that to happen, and it will guarantee it.
Mainframes are NOT the answer to all questions. Intel is NOT the answer to all questions. Itanium, Solaris, Power, etc...none are the answer to all questions.
Buy the right tool for the right purpose.
Re:Coming from Microsoft ... (Score:2, Informative)
Re:Buh buh but.... (Score:2, Informative)
By the time you migrate off a mainframe to Servers, buy lots of Microslop licences, Buy VMWare, buy Citrix, and a gaggle of utilities and backup and firewall software - the 'savings' evaporate. Factor in reliability and true recovery times - mainframe is looking great. PC Servers do not have the hardware assist of mainframes - yet.
Mainframes are still doing well, because software and OS prices have not fallen, if anything gone up. MS does not like this, because Open Source starts to look respectable and reliable and viable.And when IBM looses a sale, the salesman can just fold and say' OK if you buy it, you can run anything you like on it'
Re:I Don't Quite Understand (Score:5, Informative)
I'm not entirely certain of this but it sounds like the separation between mainframes and servers is essentially that IBM produce servers that are backwards-compatible with their very ancient mainframes. I'm not sure that in the hardware there's any specific borderline between server and mainframe. From my own experience, a LOT of companies are still using ancient COBOL-era software to run their core business. It's been around for a long time and so the bugs are ironed out and it runs OK. Software doesn't rust, and there isn't a compelling reason to replace something that works OK. However, the hardware does rust and so at some point companies need to buy hardware that will run these ancient applications.
Sounds to me like IBM is reaping the rewards of continuing to support the stuff they did 30+ years ago. The high cost with switching to another platform is rewriting their old and business critical applications. And of course reluctance to do this means accepting a very high cost of new hardware, relative to other options.
I write applications, mainly on AIX as my day job, and the hardware is very expensive, but it's not uncommon for places to still have the same servers in place 10 or 20 years down the line. It's quite common 3-4 years after an installation has gone live to have the customers IT personnel on the phone asking about replacing the hardware, and generally the advice is that there's no need to. The cost can be triple the cost of mainstream hardware, but so is the lifespan, so I think on TCO terms, it's not that bad.
NB: the stuff we write is portable C so we're not tied in to AIX in any way - my current project is running on SLES 10 on cheap Dell servers. But the real expensive servers made by IBM / Sun / HP do seem to have a reliability factor that isn't matched by cheap hardware.
Re:I Don't Quite Understand (Score:2, Informative)
Mainframes can run webservers and Linux (and specialized chips to speed Linux up) for instance. Someone needs a new LINUX server set up? Get it in minutes. The advantage they have over PC based servers is massive IO capability and uptime. And if you're using databases this is a killer speed advantage in the server world. My mainframe hasn't been shutdown in years. And as far as the OS goes, it was open source years ago, but I don't think z/OS is now. Besides, if it just works, why would any company change?
How can you claim anticompetitive? (Score:5, Informative)
Has microsoft ever had a mainframe? No.
Do they have a mainframe OS? no.
Could they develop one? HPC could theoretically be considdered one if they added storage virtualization to it, and a few other mainframe class systems.
Would we use a microsoft OS to replace out IBM mainframes? No. I'll elaborate:
- We have MILLIONS of lines of code ON the mainframe that would ALL have to be completely re-done from scratch to move off the OS390 platform.
- We have 10 times that much code that would have to be modified to talk to a non-OS390 mainframe.
- We have hundreds of servers that run support applications for the mainframe or mainframe apps that don't run on Windows.
- Any competing platform uses far more space and many fold more power, and does not have the HA features of true mainframes.
- A LARGE part of the security of our mainframe environment is that since you can't exactly get access to OS390 easily, hacing it is damned near impossible... Moving to a windows kernel based mainframe would NOT be adviseable even if we could afford it.
- IBM is here, and has been for decades, and there's more legacy code running on OS390 that's 10 years old than code running on it that's less than 10 years old. they're NOT going to drop support for it. I can't say that about any competitor.
- IBM has a FULL suite of tools to manage, monitor, and protect the mainframe. Most technologies entering the x64 space now have been in use on mainframes for 5-10 years... some longer.
- Licensing prices on the mainframe are a FRACTION of the price of lecensing x86 and P6 systems. (we're saving about 10 million this year in licensing alone moving a few hundred machines to Suse Linux virtualised on z10 IFL processors.)
- Component hardware costs of the mainframe are a bit higher (about $8K for a gig of RAM), but the system as a whole is actually not only cheaper than an equivalent VMWare or hypervisor supercluster, but it;s energy use is also a fraction of the equivalent.
- the Z systems have 5-10 year lifespans, we have a few running 12 years without a critical outage, not 3-5 years like all other platforms...
We pay a never-ending maintenance plan on our mainframes. We add new ones every year or two to replace old ones, but we don't really "buy" new mainframes, we simply pay to have a base number of MIPS available and IBM keeps the hardware running. (and pay to increase those MIPS as necessary. The licensing and hardware costs are FAR lower than out other platforms.
Re:I Don't Quite Understand (Score:4, Informative)
Power consumption? The amount of energy of having dozens of "smaller" machines, each with their separate power supplies, hard disks, RAM, etc., take, vs a similarly powerful mainframe is going to be significant.
Load balancing. I know, you said "properly load [...] balanced." But how much effort is it to properly load-balance your server farm? What if one system suddenly needs two more cores than it has? If it's "properly written" it may be able to send its work to another machine. But that means you need to maintain the software on all machines (maybe you only wanted to have it on one machine in the farm to ease sysadmin maintenance). And the cost of that software probably goes way up as they figure out how to properly partition across machines (which is a whole other beast vs merely partitioning across threads in one machines). In the mainframe world, everything is virtual, including CPUs. Need more CPU power on one machine? It'll rebalance to take that away from machines that may not be using all of their allotted CPUs. Same goes for RAM. (IIRC, you can also turn this off to give hard limits on CPU/RAM usage.)
Hot swapping, upgrading, etc. No need to take down a virtual machine just because you're replacing its CPU or RAM. Rebalancing from adding a new CPU (or set of CPUs - I don't think they come singly) is also easier. You can create a new virtual machine to use the CPU(s) or simply put it in the pool for all VMs to use. A CPU goes bad? The mainframe can take it offline without actually taking down the VM. If a server in your server farm goes down, it's just down.
I think IBM's big pushes for their mainframes come from: a) power consumption ("go green" - to hell with that, look at the money I'm saving on my electricity bills!), b) TCO (admins may cost more, but you need far fewer of them to administer a mainframe than a cluster of servers (whether AIX, Linux, HP, Solaris, or whatever), c) ease of upgrade (usually, mainframes come with a bunch of CPUs turned off and IBM doesn't charge you for them - but when you need to grow, whip out the credit card and IBM will tell you how to enable them - same-day upgrade, and you're already using the extra power - no getting a rack unit out of a box, finding space for it, putting it in, fighting with wiring, installing your software, etc., to take a week from requirement to deployment), and d) space savings ("the most expensive server you buy is the one that causes you to build a new data center" - mainframes pack more power into less space, meaning fewer physical datacenters, better climate control, etc.).
Re:How can you claim anticompetitive? (Score:3, Informative)
And that's how IBM is anticompetitive! You are "locked in" to using their hardware and software. Totally unlike MS, where you can run their software on any number of vendor's machines (HP, Dell, Gateway, even "white box" off the Internet!). And should you decide that you don't want to run MS Office any more, why then it is simple to convert to ... OOPS - never mind. Or if you want to integrate a non-Windows server into an Active Directory environment, you simply ... Never mind again. Or remember how easy it is to run a Win95 app on Vista - DAMN! forget that one too.
For the slow, the above is sarcasm. Not at its finest, granted.
--
John
Re:They could be right even when they're doing wro (Score:5, Informative)
Microsoft has plenty of money. If they don't like the way the mainframe market looks then they should enter and build their own. IBM has already been through the anti-trust wringer for their mainframe hardware and has spent decades under supervision by the Justice Department.
The article is missing the fact that T3 bought it's technology from Platform Solutions. Platform Solutions was acquired [ibm.com] by IBM. Without reselling Platform Solutions' product I don't see how T3 has any offerings that IBM competes with. They look like a distributor that has been cut off form a supplier, that's not grounds for anti-trust.
Re:I Don't Quite Understand (Score:3, Informative)
No, they're still playing "Air Traffic Control" (Score:2, Informative)
http://www.faa.gov/about/office_org/headquarters_offices/ato/publications/oep/version1/reference/eram/ [faa.gov] Host Computer System is a G3 mainframe running code from the 1960's/70's, although the FAA is on-schedule to replace it by 2012 with non-mainframe computers.
Now, *that's* a "business critical" application!