If you want to make money in the mainframe racket, you have to keep building ever more capacious machines. And if you really want to make money, then you port your code from CMOS mainframe engines to Intel-based machinery and stop making homegrown processors and system boards. Unisys is not quite there yet – it's about halfway …
Civil Service Imperatives
> However, Unisys sells a lot of its mainframes to federal and state governments, and among these customers, budgets are set and they don't care a whit about utility pricing even if it might save money. (Your tax dollars, pounds, and euros at rest.)
Please be a little more charitable to your hard-working public employees. The guys and gals putting together the recommendation will *never* be rewarded for proposing an efficient solution that involves an uncertain cost -- that's gambling, innit?
Very nice to Unisis
Why didn't TPM rip them apart, the same way he rips apart Oracle, and formerly Sun?
Not even a sarcastic comment about a really slow speed bump...
Not that this is bad, it would be nice for TPM to show the same level off even-handedness with Oracle/Sun!
maybe because Oracle is killing the Sun gear and not being very truthful
There is a world of difference between what Unisys is doing FOR their install base and what Oracle is doing TO their install base
check out the following for more details
I'm always interested in how these machines configure the gigabytes of main storage. The OS2200 boxes were (are?) 36 bit word machines and the MCP boxes were (are?) 48 bit words with lots of additional typing bits per word. I think the original MCP machines from the 1960's (B5500) had 53 bit words, 48 bits of content and 5 bits of descriptor, there were invisible ECC bits too. It could have been a later (1970's) version (B6700) with that config.
Also, how the heck do Xeon processors provide either of these environments? Straight emulation/simulation? The entry machines must spend most of their time decoding and shifting bits about.
Worked on both both systems a bit in the 1980's. When Unisys was formed the general view of the employees was it would be a disaster. Within a few years the joke was that it demonstrated how to take two $2billion companies and make one $2billion company.
Good fun just the same.
"...Also, how the heck do Xeon processors provide either of these environments? Straight emulation/simulation? The entry machines must spend most of their time decoding and shifting bits about...."
If these Mainframes are similar to IBM Mainframes, then it should be no problem to emulate them The TurboHercules emulator for x86, can give 3.200 MIPS on a 8-socket Nehalem box. The biggest IBM mainframe is 10x-15x faster. That is not extremely good a result from IBM, considering that IBM Mainframes run native code, and the x86 runs software emulation which is 5-10x slower. Also, the IBM Mainframe has 20 cpus vs the 8-socket Nehalem box.
In short, the x86 should be able to emulate Mainframes very fast. If you ported the software to x86, you would get 5-10x higher performance.