| Author |
Message |
Del Cecchi
Guest
|
Posted:
Mon Feb 07, 2005 11:19 pm Post subject:
Cell press release, redacted. |
|
|
IBM, Sony, Sony Computer Entertainment Inc. and Toshiba Disclose Key
Details of the Cell Chip
Monday February 7, 1:00 pm ET
Innovative Design Features Eight Synergistic Cores Together with Power
Based Core, Delivers More Than 10 Times the Performance of the Latest PC
Processors
SAN FRANCISCO--(BUSINESS WIRE)--Feb. 7, 2005-- At the International
Solid State Circuits Conference (ISSCC) today, IBM, Sony Corporation,
Sony Computer Entertainment Inc. (Sony and Sony Computer Entertainment
collectively referred to as Sony Group) and Toshiba Corporation
(Toshiba) for the first time disclosed in detail the breakthrough
multi-core architectural design - featuring supercomputer-like floating
point performance with observed clock speeds greater than 4 GHz - of
their jointly developed microprocessor code-named Cell.
A team of IBM, Sony Group and Toshiba engineers has collaborated on
development of the Cell microprocessor at a joint design center
established in Austin, Texas, since March 2001. The prototype chip is
221 mm(2), integrates 234 million transistors, and is fabricated with 90
nanometer SOI technology.
(snip)
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
Initial production of Cell microprocessors is expected to begin at IBM's
300mm wafer fabrication facility in East Fishkill, N.Y., followed by
Sony Group's Nagasaki Fab, this year. IBM, Sony Group and Toshiba expect
to promote Cell-based products including a broad range of industry-wide
applications, from digital televisions to home servers to supercomputers.
Among the highlights of Cell released today:
* Cell is a breakthrough architectural design -- featuring eight
synergistic processors and top clock speeds of greater than 4 GHz (as
measured during initial hardware testing)
* Cell is a multicore chip capable of massive floating point processing
* Cell is OS neutral and supports multiple operating systems
simultaneously
(snip)
-----------------not much there when the fluff is gone------------------- |
|
| Back to top |
|
 |
David Wang
Guest
|
Posted:
Tue Feb 08, 2005 12:10 am Post subject:
Re: Cell press release, redacted. |
|
|
Del Cecchi <cecchinospam@us.ibm.com> wrote:
| Quote: | Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
|
I just came out of the SPE session, and I have to say that I don't know
what this paragraph above means.
Support all the multiple OS's and real time OS's and guest OS's...
simultaneously?
That doesn't sound right.
There's a 64b PPC processor, in order, dual threaded, and that's the
master CPU, so you can run any OS you want, but the SPE's aren't
"real CPUs", as far as what we generally want in a "real CPU"
these days. Specifically, quoting directly from the ISSCC paper
on the SPE's, it states
.. . . "The focus on efficiency comes at the cost of multi-user operating
system support. SPE load and store operations are performed within a
local address space, not in system address space." . . .
I am befuddled.
--
davewang202(at)yahoo(dot)com |
|
| Back to top |
|
 |
Ketil Malde
Guest
|
Posted:
Tue Feb 08, 2005 1:56 am Post subject:
Re: Cell press release, redacted. |
|
|
Del Cecchi <cecchinospam@us.ibm.com> writes:
| Quote: | * Cell is a breakthrough architectural design -- featuring eight
synergistic processors
|
Does that mean the OS (interrupt handling?) run on one core, and the
eight "synergistic" ones provide (user space) processing power, or
something like that?
| Quote: | and top clock speeds of greater than 4 GHz
|
So long, P4! Power consumption seems left out, too technical for a
press release?
| Quote: | (as measured during initial hardware testing)
|
What does this mean?
| Quote: | * Cell is OS neutral and supports multiple operating systems
simultaneously
|
Virtualization, in other words? (Of course, the OS most people need to
run occasionally, is limited to a different architecture...)
From a similar posting:
| broadband processor...industrial applications...consumer electronics
| applications...digital televisions
The chip sounds nice enough (I guess they all do in press releases),
but am I the only one who find it odd that here we have this
super-incredible chip, and IBM and Toshiba seem to be aiming it
squarely at the embedded controller market? :-)
-kzm
--
If I haven't seen further, it is by standing in the footprints of giants |
|
| Back to top |
|
 |
Joseph Seigh
Guest
|
Posted:
Tue Feb 08, 2005 2:06 am Post subject:
Re: Cell press release, redacted. |
|
|
On Mon, 07 Feb 2005 12:19:19 -0600, Del Cecchi <cecchinospam@us.ibm.com> wrote:
| Quote: | (snip)
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
|
[...]
| Quote: |
Among the highlights of Cell released today:
* Cell is a breakthrough architectural design -- featuring eight
synergistic processors and top clock speeds of greater than 4 GHz (as
measured during initial hardware testing)
* Cell is a multicore chip capable of massive floating point processing
* Cell is OS neutral and supports multiple operating systems
simultaneously
|
I assume being OS neutral supporting multiple guest operating systems
simutaneously means the architecture virualizes without having to hack
guest OSes. ...apart from having to port the OSes to the new processor
in the first place.
--
Joe Seigh |
|
| Back to top |
|
 |
David Wang
Guest
|
Posted:
Tue Feb 08, 2005 2:43 am Post subject:
Re: Cell press release, redacted. |
|
|
del cecchi <dcecchi.nojunk@att.net> wrote:
| Quote: | "David Wang" <foo@bar.invalid> wrote in message
Del Cecchi <cecchinospam@us.ibm.com> wrote:
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
I just came out of the SPE session, and I have to say that I don't know
what this paragraph above means.
Support all the multiple OS's and real time OS's and guest OS's...
simultaneously?
Make that "simultaneously". I presume that refers to the virtualization
that was mentioned in one of the press releases. So all those OSs can
run on the main PowerPC. It's what the PowerPC servers do.
|
Right. That's what it was. Through a Hypervisor above the OS,
dynamically allocate the resources, but everything is based off
of the PPE. The SPE's are just another set of resources that can
be utilized.
--
davewang202(at)yahoo(dot)com |
|
| Back to top |
|
 |
Kai Harrekilde-Petersen
Guest
|
Posted:
Tue Feb 08, 2005 3:07 am Post subject:
Re: Cell press release, redacted. |
|
|
Ketil Malde <ketil+news@ii.uib.no> writes:
| Quote: | Del Cecchi <cecchinospam@us.ibm.com> writes:
and top clock speeds of greater than 4 GHz
So long, P4! Power consumption seems left out, too technical for a
press release?
(as measured during initial hardware testing)
What does this mean?
|
<tongue in cheek>
Single part(s), tweaked like h*** with increased power supplies, extra
cooling (Liquid Nitrogen is non-conducting) etc.
</tongue in cheek>
Kai
--
Kai Harrekilde-Petersen <khp(at)harrekilde(dot)dk> |
|
| Back to top |
|
 |
Arrvindh Shriraman
Guest
|
Posted:
Tue Feb 08, 2005 3:54 am Post subject:
Re: Cell press release, redacted. |
|
|
Joseph Seigh wrote:
| Quote: | On Mon, 07 Feb 2005 12:19:19 -0600, Del Cecchi <cecchinospam@us.ibm.com
wrote:
(snip)
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
[...]
Among the highlights of Cell released today:
* Cell is a breakthrough architectural design -- featuring eight
synergistic processors and top clock speeds of greater than 4 GHz (as
measured during initial hardware testing)
* Cell is a multicore chip capable of massive floating point
processing
* Cell is OS neutral and supports multiple operating systems
simultaneously
I assume being OS neutral supporting multiple guest operating systems
simutaneously means the architecture virualizes without having to hack
guest OSes. ...apart from having to port the OSes to the new processor
in the first place.
|
I dont see how this is different from intel's own Vanderpool technology.
Also Sony has fallen short to deliver on 65nm technology. Intel still i
guess rules the technology roost afterall. HAs anyone got an actual copy
for the presentation. Who said architecture research is dead ?
Its alive and kicking ass
Cheers |
|
| Back to top |
|
 |
Niels Jørgen Kruse
Guest
|
Posted:
Tue Feb 08, 2005 3:57 am Post subject:
Re: Cell press release, redacted. |
|
|
Ketil Malde <ketil+news@ii.uib.no> wrote:
| Quote: | Del Cecchi <cecchinospam@us.ibm.com> writes:
and top clock speeds of greater than 4 GHz
So long, P4! Power consumption seems left out, too technical for a
press release?
|
This was posted at RWT by David Wang:
| Quote: | Today is the day that the CELL processor family gets announced.
I'll be writing a few things about the various papers at ISSCC on CELL,
but before the news conference starts and the papers gets official
unveiling, some interesting data have already been presented in the
technical digests.
The CELL processor presented has 1 64b PPC core acting as the traditional
scalar processor, complete with its own L2. The PPE (PowerPC processing
Element) is connected to 8 other SPE's (Synergistic Processing Elements)
The SPE's are the magic glue that is suppose to contain enormous amount of
compute power and a bunch of them gets you the enormously large flop
rating that we've all head much about.
Some stats.
1. 90nm SOI process.
2. Logic depth is functionally equivalent to about 20 FO4 (est), but
circuit speed equivalence is 11 FO4 per stage. The short pipestage circuit
depth is reached with "circuit efficiencies" and Dynamic logic !?!
3. With per stage delay of 11 FO4, the schmoo plots show that the SPE's
can crank from 3.2 GHz @ 0.9V Vdd to 5.2 GHz @ 1.3 V Vdd. The entire chip
has similar frequency/voltage range, but to get to 5.2 Ghz @ 1.3V, each
SPE will eat 11~12W. Add in the rest and the chip will get really hot. 4
GHz @ 1.1V = 4W per SPE seems to be the nominal range.
4. Die size per SPE is 2.5 x 5.81 mm^2. The entire chip with 8 SPE's seems
to be about 17.2 x 12 mm^2. That seems to be an awfully large chip for
IBM. The CPU to be used in PS3/Xbox2 will probably be the 65nm version or
it'll have to have fewer SPE's.
6. As previously announced, the off chip I/O interface is Rambus Redwood
and the memory interface is XDR. Similar clocking/deskewing schemes. Looks
to be about ~50 GB/s BW to memory, and 50~100 GB/s to I/O.
I'll write up articles as the papers are presented.
|
--
Mvh./Regards, Niels Jørgen Kruse, Vanløse, Denmark |
|
| Back to top |
|
 |
Del Cecchi
Guest
|
Posted:
Tue Feb 08, 2005 4:00 am Post subject:
Re: Cell press release, redacted. |
|
|
Ketil Malde wrote:
| Quote: | Del Cecchi <cecchinospam@us.ibm.com> writes:
* Cell is a breakthrough architectural design -- featuring eight
synergistic processors
Does that mean the OS (interrupt handling?) run on one core, and the
eight "synergistic" ones provide (user space) processing power, or
something like that?
If I understood the press release, there is a powerPC that runs the OS, |
and 8 other processors that are analogous to GPU (not GPU, but sort of
the coprocessor idea)
| Quote: |
and top clock speeds of greater than 4 GHz
So long, P4! Power consumption seems left out, too technical for a
press release?
|
Probably quoted in the ISSCC Presentation. Watch EEtimes, EDN,
Electronic News, and Electronic Design (all online magazines in addition
to hard copy) for reports.
| Quote: |
(as measured during initial hardware testing)
What does this mean?
|
They have one running in the lab. At least one. Not commenting on
distribution of frequency. Probably not using heroic overclocking.
| Quote: |
* Cell is OS neutral and supports multiple operating systems
simultaneously
Virtualization, in other words? (Of course, the OS most people need to
run occasionally, is limited to a different architecture...)
|
Linux? Nah, run linux on powerPC. Oh, you mean Windows? Why would you
run Windows on this?
| Quote: |
From a similar posting:
| broadband processor...industrial applications...consumer electronics
| applications...digital televisions
The chip sounds nice enough (I guess they all do in press releases),
but am I the only one who find it odd that here we have this
super-incredible chip, and IBM and Toshiba seem to be aiming it
squarely at the embedded controller market? :-)
|
IBM Toshiba and Sony are aiming it at the Game and Home market, seems to
me.
|
|
| Back to top |
|
 |
Joseph Seigh
Guest
|
Posted:
Tue Feb 08, 2005 5:43 am Post subject:
Re: Cell press release, redacted. |
|
|
On Tue, 8 Feb 2005 00:10:07 +0000 (UTC), David Wang <foo@bar.invalid> wrote:
| Quote: | Del Cecchi <cecchinospam@us.ibm.com> wrote:
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
I just came out of the SPE session, and I have to say that I don't know
what this paragraph above means.
Support all the multiple OS's and real time OS's and guest OS's...
simultaneously?
That doesn't sound right.
There's a 64b PPC processor, in order, dual threaded, and that's the
master CPU, so you can run any OS you want, but the SPE's aren't
"real CPUs", as far as what we generally want in a "real CPU"
these days. Specifically, quoting directly from the ISSCC paper
on the SPE's, it states
. . . "The focus on efficiency comes at the cost of multi-user operating
system support. SPE load and store operations are performed within a
local address space, not in system address space." . . .
I am befuddled.
There's some info here |
http://www.scee.presscentre.com/imagelibrary/detail.asp?MediaDetailsID=25555
- Contains 8 SPUs each containing a 128 entry 128-bit register file and
256KB Local Store
The local stuff sounds like numa
- Contains 64-bit Power ArchitectureTM with VMX that is a dual thread SMT design
- views system memory as a 10-way coherent threaded machine
make that ccNuma
--
Joe Seigh |
|
| Back to top |
|
 |
del cecchi
Guest
|
Posted:
Tue Feb 08, 2005 6:44 am Post subject:
Re: Cell press release, redacted. |
|
|
"David Wang" <foo@bar.invalid> wrote in message
news:cu900v$10d$1@grapevine.wam.umd.edu...
| Quote: | Del Cecchi <cecchinospam@us.ibm.com> wrote:
Another advantage of Cell is to support multiple operating systems,
such
as conventional operating systems (including Linux), real-time
operating
systems for computer entertainment and consumer electronics
applications
as well as guest operating systems for specific applications,
simultaneously.
I just came out of the SPE session, and I have to say that I don't
know
what this paragraph above means.
Support all the multiple OS's and real time OS's and guest OS's...
simultaneously?
That doesn't sound right.
|
Make that "simultaneously". I presume that refers to the virtualization
that was mentioned in one of the press releases. So all those OSs can
run on the main PowerPC. It's what the PowerPC servers do.
| Quote: |
There's a 64b PPC processor, in order, dual threaded, and that's the
master CPU, so you can run any OS you want, but the SPE's aren't
"real CPUs", as far as what we generally want in a "real CPU"
these days. Specifically, quoting directly from the ISSCC paper
on the SPE's, it states
. . . "The focus on efficiency comes at the cost of multi-user
operating
system support. SPE load and store operations are performed within a
local address space, not in system address space." . . .
I am befuddled.
--
davewang202(at)yahoo(dot)com
|
Having the advantage of knowing nothing about the chip except what I
read in the press, I would submit that it has a full function
virtualized lpar'd PowerPC with L1, L2, memory all that. In addition it
has 8 of these special purpose attached processors, analogous to an 8087
or an IOP or some other specialized attached processor. Each attached
processor gets to own a chunk of memory. Since it seems unlikely that
physical memory is disjoint, the physical address spaces for all 8 must
be disjoint. So they each have "their own", perhaps shared with the
main processor in some way. Or perhaps not.
del cecchi |
|
| Back to top |
|
 |
Patrick Schaaf
Guest
|
Posted:
Tue Feb 08, 2005 7:58 am Post subject:
Re: Cell press release, redacted. |
|
|
Arrvindh Shriraman <arrvindh_shriraman@yahoo.com> writes:
| Quote: | * Cell is OS neutral and supports multiple operating systems
simultaneously
I dont see how this is different from intel's own Vanderpool technology.
|
Interesting. Can you expound a bit on the common technical components
of those two technologies? What exactly do they do that current processors
do not?
All press releases I've seen have been very vague on details. I'd love
to be informed.
| Quote: | Also Sony has fallen short to deliver on 65nm technology.
|
Oh, is 65nm technology already delivered by anybody? You seem to imply Intel:
which Intel processors currently deliver using 65nm technology?
curious
Patrick |
|
| Back to top |
|
 |
Arrvindh Shriraman
Guest
|
Posted:
Tue Feb 08, 2005 7:58 am Post subject:
Re: Cell press release, redacted. |
|
|
Pls take a look at intel's Vanderpool technology
http://www.intel.com/technology/computing/vptech/index.htm
IBM i guess beat intel to this one.
But i am still betting on intel beating IBM to 65nm. BTW has anybody got
hold of the paper or presentation. It would be good to have a look at
the Cell's architecture and technological similarities between the Blue
Gene's node and Cell. So how were the other presentations at ISSCC
anything other than Cell which attracted people
Del Cecchi wrote:
| Quote: | IBM, Sony, Sony Computer Entertainment Inc. and Toshiba Disclose Key
Details of the Cell Chip
Monday February 7, 1:00 pm ET
Innovative Design Features Eight Synergistic Cores Together with Power
Based Core, Delivers More Than 10 Times the Performance of the Latest PC
Processors
SAN FRANCISCO--(BUSINESS WIRE)--Feb. 7, 2005-- At the International
Solid State Circuits Conference (ISSCC) today, IBM, Sony Corporation,
Sony Computer Entertainment Inc. (Sony and Sony Computer Entertainment
collectively referred to as Sony Group) and Toshiba Corporation
(Toshiba) for the first time disclosed in detail the breakthrough
multi-core architectural design - featuring supercomputer-like floating
point performance with observed clock speeds greater than 4 GHz - of
their jointly developed microprocessor code-named Cell.
A team of IBM, Sony Group and Toshiba engineers has collaborated on
development of the Cell microprocessor at a joint design center
established in Austin, Texas, since March 2001. The prototype chip is
221 mm(2), integrates 234 million transistors, and is fabricated with 90
nanometer SOI technology.
(snip)
Another advantage of Cell is to support multiple operating systems, such
as conventional operating systems (including Linux), real-time operating
systems for computer entertainment and consumer electronics applications
as well as guest operating systems for specific applications,
simultaneously.
Initial production of Cell microprocessors is expected to begin at IBM's
300mm wafer fabrication facility in East Fishkill, N.Y., followed by
Sony Group's Nagasaki Fab, this year. IBM, Sony Group and Toshiba expect
to promote Cell-based products including a broad range of industry-wide
applications, from digital televisions to home servers to supercomputers.
Among the highlights of Cell released today:
* Cell is a breakthrough architectural design -- featuring eight
synergistic processors and top clock speeds of greater than 4 GHz (as
measured during initial hardware testing)
* Cell is a multicore chip capable of massive floating point processing
* Cell is OS neutral and supports multiple operating systems
simultaneously
(snip)
-----------------not much there when the fluff is gone-------------------
|
|
|
| Back to top |
|
 |
Ketil Malde
Guest
|
Posted:
Tue Feb 08, 2005 1:03 pm Post subject:
Re: Cell press release, redacted. |
|
|
Del Cecchi <cecchinospam@us.ibm.com> writes:
| Quote: | If I understood the press release, there is a powerPC that runs the
OS, and 8 other processors that are analogous to GPU (not GPU, but
sort of the coprocessor idea)
|
That was my impression, too. I know this has been discussed on
c.a. at regular intervals, but is it a first in silicon? That is, if
the "synergistic" processors can schedule normal user code?
| Quote: | What does this mean?
They have one running in the lab. At least one. Not commenting on
distribution of frequency. Probably not using heroic overclocking.
|
Probably just enough to get that "4" to spite Intel? :-)
| Quote: | Virtualization, in other words? (Of course, the OS most people need
to run occasionally, is limited to a different architecture...)
Linux? Nah, run linux on powerPC.
|
Exactly. I don't need (although it could be a benefit) virtualization
to run Linux, since it can be modified to run on a multiplexing system
(UML or Xen, for instance).
| Quote: | Why would you run Windows on this?
|
Because my bank, which apparently is run by crack-smoking monkeys,
decided that the old, working solution was too good for me, and I now
will need IE to pay my bills?
More generally, people like VMWare make a living mainly from providing
Windows to non-Windows users. Partly Citrix and others as well.
-kzm
--
If I haven't seen further, it is by standing in the footprints of giants |
|
| Back to top |
|
 |
Andrew Reilly
Guest
|
Posted:
Tue Feb 08, 2005 4:41 pm Post subject:
intel's Vanderpool and virtualization in general (was Re: Ce |
|
|
On Tue, 08 Feb 2005 03:51:22 -0500, Arrvindh Shriraman wrote:
Now, I can see that this is pretty much a reaction to the seemingly large
success of things like VMWare and Microsoft's VirtualPC (on PC).
Finding ways to optimize popular software is always a good idea.
Also notwithstanding that there's an obvious pre-existing example of
wholesale virtualization in the IBM mainframes, is there actually a really
good reason to use virtualization on a day-to-day basis, or is it just a
practical acknowledement that there are failings of the OSes that are used
on these things?
I mean, OSes are supposed to be there to ration access to a machine's
hardware resources, for the benefits of the (several) applications that
want to share it.
What do virtualizers provide that couldn't be incorporated into an OS?
(Save on porting effort for applications coded against a different API?
Save some memory space, compared to the various JIT/Dynamic recompilation
alternatives?)
Cheers,
--
Andrew |
|
| Back to top |
|
 |
|
|
|
|