Re: Is a RISC chip more expensive?
- From: Stephen Fuld <S.Fuld@xxxxxxxxxxxxxxxxxxxx>
- Date: Fri, 31 Aug 2007 03:39:27 GMT
Quadibloc wrote:
Stephen Fuld wrote:Quadibloc wrote:
I read a recent news item to the effect that Univac has abandonedCan you point me to a source for that? As an old Univac systems
making custom CPU chips for *both* the 1100 descendants and the A-
Series descendants, and are now using an emulation layer on x86 for
*all* new computer designs.
programmer, I would really like to see what they are doing. Thanks.
Here's one:
Thanks, John. But I don't think these mean what you think they mean. see below.
http://www.unispheremag.com/5_minute_archives/7-06.html
Not only does it use "Intel Xeon and Itanium processor technologies",
and run "Unisys's own OS2200 and MCP", but the article does state
"This next-generation architecture will also serve as the basis for
all Unisys enterprise servers"
Also,
http://m-net.net.nz/1653/databases/news/unisys-unveils-next-gen-common-architecture-mainframes.php
For some time, Unisys has offered systems composed of multiple processor types. They consist of one or more Intel Xeon processors and one or more of their own proprietary CPUs. These all share the same memory that is statically partitioned between them. There is a high speed link between the proprietary CPUs and the Xeons to allow fast communication between them. The idea is to allow the customer to maintain its investment in the applications programs, database technology, etc. from the proprietary systems and "marry" them to the much better presentation tools, and front end logic offered on the Xeon systems. For example, you could run a web server on the Xeon that got its data from a database residing on the proprietary system.
I believe the systems that these references talk about are the next generation of that type of system. A few quotations to bolster my position.
From the first one
The new Dorado and Libra models permit Intel and *CMOS processors* to co-exist in a dynamically provisioned environment. They also feature partitioning capabilities for balancing application workloads across different environments.
(Added emphasis mine)
The use of the term CMOS is "Unisys speak" for their proprietary processors (1100 series and A- series) that made the transition from bi polar to CMOS some years ago.
And from the second one
These releases provide support for ClearPath systems using Intel or *proprietary processor technology* and both operating environments provide full compatibility for applications written on earlier ClearPath systems based on *CMOS processors*.
(Again, emphasis is mine). It is clearer here. They are providing both Intel and their proprietary processors within the same system (different partitions).
--
- Stephen Fuld
(e-mail address disguised to prevent spam)
.
- Follow-Ups:
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Re: Is a RISC chip more expensive?
- References:
- Is a RISC chip more expensive?
- From: mike3
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Re: Is a RISC chip more expensive?
- From: mike3
- Re: Is a RISC chip more expensive?
- From: Florian Laws
- Re: Is a RISC chip more expensive?
- From: Del Cecchi
- Re: Is a RISC chip more expensive?
- From: Florian Laws
- Re: Is a RISC chip more expensive?
- From: Del Cecchi
- Re: Is a RISC chip more expensive?
- From: Nick Maclaren
- Re: Is a RISC chip more expensive?
- From: Stephen Fuld
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Re: Is a RISC chip more expensive?
- From: Stephen Fuld
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Re: Is a RISC chip more expensive?
- From: Stephen Fuld
- Re: Is a RISC chip more expensive?
- From: Quadibloc
- Is a RISC chip more expensive?
- Prev by Date: Re: Is a RISC chip more expensive?
- Next by Date: Re: Is a RISC chip more expensive?
- Previous by thread: Re: Is a RISC chip more expensive?
- Next by thread: Re: Is a RISC chip more expensive?
- Index(es):
Relevant Pages
|