|
Tuesday December 06, 2005
Ultrasparc-T1 T2000 I was going to compare one of the new UltraSparc-T1 machines to
some other boxes, but as there is nothing of a similar size,
packing a similar punch, anywhere on the planet I figured
let's let the T2000 speak for itself........
# uname -a
SunOS oaf251 5.11 snv_27 sun4v sparc SUNW,Sun-Fire-T2000
A view from the front
Occasionally a picture is not worth 1000 words[1], or 32 cpu threads...
# prtdiag
System Configuration: Sun Microsystems sun4v Sun Fire T2000
System clock frequency: 200 MHz
Memory size: 16376 Megabytes
========================= CPUs ===============================================
CPU CPU
Location CPU Freq Implementation Mask
------------ ----- -------- ------------------- -----
MB/CMP0/P0 0 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P1 1 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P2 2 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P3 3 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P4 4 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P5 5 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P6 6 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P7 7 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P8 8 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P9 9 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P10 10 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P11 11 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P12 12 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P13 13 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P14 14 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P15 15 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P16 16 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P17 17 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P18 18 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P19 19 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P20 20 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P21 21 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P22 22 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P23 23 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P24 24 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P25 25 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P26 26 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P27 27 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P28 28 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P29 29 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P30 30 1200 MHz SUNW,UltraSPARC-T1
MB/CMP0/P31 31 1200 MHz SUNW,UltraSPARC-T1
========================= IO Configuration =========================
IO
Location Type Slot Path Name Model
----------- ----- ---- --------------------------------------------- ------------------------- ---------
IOBD/NET0 PCIE IOBD /pci@780/pci@0/pci@1/network@0 network-pciex8086,105e
IOBD/NET-1 PCIE IOBD /pci@780/pci@0/pci@1/network network-pciex8086,105e
IOBD/PCIX PCIX IOBD /pci@7c0/pci@0/pci@1/pci@0/isa@2 isa
IOBD/PCIX PCIX IOBD /pci@7c0/pci@0/pci@1/pci@0/usb@5 usb-pciclass,0c0310
IOBD/PCIX PCIX IOBD /pci@7c0/pci@0/pci@1/pci@0/usb@6 usb-pciclass,0c0310
IOBD/PCIX PCIX IOBD /pci@7c0/pci@0/pci@1/pci@0/ide ide-pci10b9,5229
IOBD/PCIX PCIX PCIX /pci@7c0/pci@0/pci@1/pci@0,2/LSILogic,sas@2 LSILogic,sas-pci1000,50 LSI,1064
IOBD/NET-1 PCIE IOBD /pci@7c0/pci@0/pci/network network-pciex8086,105e
IOBD/NET-1 PCIE IOBD /pci@7c0/pci@0/pci/network network-pciex8086,105e
But is that just one cpu I see?
Yes, yes it is.
And yes that is 32 virtual CPUs you see.....
# psrinfo -p -v
The physical processor has 32 virtual processors (0, 1, 2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)
UltraSPARC-T1 (cpuid 0 clock 1200 MHz)
# psrinfo
0 on-line since 12/06/2005 12:03:41
1 on-line since 12/06/2005 12:03:42
2 on-line since 12/06/2005 12:03:42
3 on-line since 12/06/2005 12:03:42
4 on-line since 12/06/2005 12:03:42
5 on-line since 12/06/2005 12:03:42
6 on-line since 12/06/2005 12:03:42
7 on-line since 12/06/2005 12:03:42
8 on-line since 12/06/2005 12:03:42
9 on-line since 12/06/2005 12:03:42
10 on-line since 12/06/2005 12:03:42
11 on-line since 12/06/2005 12:03:42
12 on-line since 12/06/2005 12:03:42
13 on-line since 12/06/2005 12:03:42
14 on-line since 12/06/2005 12:03:42
15 on-line since 12/06/2005 12:03:42
16 on-line since 12/06/2005 12:03:42
17 on-line since 12/06/2005 12:03:42
18 on-line since 12/06/2005 12:03:42
19 on-line since 12/06/2005 12:03:42
20 on-line since 12/06/2005 12:03:42
21 on-line since 12/06/2005 12:03:42
22 on-line since 12/06/2005 12:03:42
23 on-line since 12/06/2005 12:03:42
24 on-line since 12/06/2005 12:03:42
25 on-line since 12/06/2005 12:03:42
26 on-line since 12/06/2005 12:03:42
27 on-line since 12/06/2005 12:03:42
28 on-line since 12/06/2005 12:03:42
29 on-line since 12/06/2005 12:03:42
30 on-line since 12/06/2005 12:03:42
31 on-line since 12/06/2005 12:03:42
Internally the box looks like this
And let's not forget the four e1000g's that are integrated as standard
Yep, I'm speechless, these are incredible machines. And they don't rhyme with Hell
[1] For the pedantic
# prtdiag | wc -w
255
(2005-12-06 08:55:00.0)
Permalink
Tuesday October 18, 2005
Some self effacing humour....
A good while ago I managed to christen myself with a nickname within my group of Grumpy, more in a reaction to being called Whiney, who is not generally known as one of the dwarfs, but does happen to be a member of my team (I'll let him own up to this himself). The original context was along the lines of someone calling my Whiney, and I replied, "thats not me, I'm grumpy". As you might guess the name stuck. Anyway Niall was over in California recently, and while wondering through a Disney store with his better half spotted an apt mug..... so I give you....
My new mugshot ;). (pun intended).
(2005-10-18 03:14:45.0)
Permalink
Wednesday September 21, 2005
workaround for "console is not shared" netra x1 lom error message
Ah, back to my "thank you for the informative error messages" postings. On a netra X1 box you may possibly get an error message such as
lom>console
Console is not shared
which is not very helpful. Of course one would think that console -f would override this, but nope, it doesn't. So the work around, which will force a reboot of your machine is the following.
lom>bootmode reset_nvram
lom>reset
lom>
LOM event: +128d+1h34m23s host reset
lom>console
oaf108 console login:
and thats it. You can also connect directly to the serial port and reset your eeprom settings there as well.
(2005-09-21 08:06:38.0)
Permalink
Monday September 12, 2005
Shiny new Galaxy's.......
Sun released a bunch of cool new boxes today, which we have been using internally for a little while now. One of the guys in my group, Niall Mullen (who joins the blogosphere today as well) has a few pics of the boxes and a nice walk through some of the cooler features up here. He spares you the press release jargon, and show you the real deal. But to put it simply these run faster, cooler and cheaper than anything else that any other vendor is making.
(2005-09-12 13:24:17.0)
Permalink
Friday August 26, 2005
A new slashdot low?
Slashdot stories can be annoying and petty at the best of times, but an article today regarding a Microsoft employee who has gotten adult measels is definately a new low. One of my colleagues had this recently - its damn serious, not something to make fun of. At least some of the folks commenting on the article see it the same way.
(2005-08-26 02:43:32.0)
Permalink
Friday July 15, 2005
semi work related....
I tend to avoid non work related stuff on this blog, but as this is impacting my ability to work from home, I guess its semi work related. I just wrote up a small piece about my current broadband fiasco over on my personal blog.
(2005-07-15 03:23:40.0)
Permalink
Tuesday May 24, 2005
Managers blog......
So our manager (or ex-engineer as we prefer to call him) has joined bsc, drum roll please....... introducing Damien Farnham. Back in his engineering days he worked on things such as TPC and the like (predominantly focused on Sybase). Or so he likes to tell us ;). These days he does all of that management stuff that us engineers like to avoid ;).
(2005-05-24 05:14:24.0)
Permalink
Wednesday May 04, 2005
uint*_t and standards
After my post about getting libnjb to compile on Solaris I got a couple of comments around uint*_t types, and why we defined uint's in the manner that we do. Mainly because the convention we use is a standard, as per Ansi C99, the Single Unix Specification (which supercedes Posix and XPG).
Unfortunately all of these specs cost money to download (something of a bone of contention with me, open standards should be free to view), but the best online explanation that I can find is available in this article by Michael Barr over on O'Reilly. From that article I quote
In hindsight, it sure would have been nice if the authors of the C standard had defined
some standard names and made compiler providers responsible for providing the appropriate
typedef for each fixed-size integer type in a library header file. Alternatively, of
course, the C standard could have specified (as Java does) that each of the types short,
int, and long has a standard width on all platforms; but that can have an impact on
performance, particularly on 8-bit processors that must implement 16- and 32-bit additions
in multi-instruction sequences.
Interestingly, it turns out that the authors of a 1999 update to the ISO C standard
(hereafter "C99") did just that. It seems the ISO organization has finally put the weight
of its standard behind a preferred set of names for signed and unsigned fixed-size integer
data types. The newly defined type names are as follows:
8-bit: int8_t uint8_t
16-bit: int16_t uint16_t
32-bit: int32_t uint32_t
64-bit: int64_t uint64_t
So there you go.
[ update - 05/05/05 ]
A couple of people have commented that you can access the Unix O3 spec here, thanks for the clarification. Anyone got a url for a C99 spec that they can point us at?
[update - 07/05/05 ]
Terry Lambert pointed me at a C99 spec.
(2005-05-03 20:29:08.0)
Permalink
Tuesday May 03, 2005
Echo Chambers?
Fascinating article over on The Register entitled SCO, Groklaw and the Monterey mystery that never was. Its a lot closer to the history that I remember around project monterey.
Whats interesting is the commentary about online forums turning into "echo chambers" (or could that be Harry Potter and the chambers of conspiracy ?). The current echo on slashdot around this story is sadly predictable.
(2005-05-02 16:29:41.0)
Permalink
Wednesday April 27, 2005
Binary compatability and application dependicies
I see Oracle are echoing what many customers have been saying about Linux patches in this article over on the register.
"Speaking at the Software 2005 conference in Santa Clara, California, on Tuesday, Phillips told an audience of 1,000 chief executive officers: "[Customers] are bogged down by the dependencies between applications. Add a patch in Linux and five other things break. People ask: 'Why can't you tell me what the dependencies are?'"
This is exactly why binary compatability and interface stability are important. If you can't add patches with the confidence that your production systems will stay up, and that your applications will stay running on them, you can only be nervous. And it is also why Sun invests a huge amount of time into testing patches.
(2005-04-26 20:47:36.0)
Permalink
Wednesday April 20, 2005
Workaround for "FATAL: system is not bootable, boot command is disabled" on an obp
Poor error messages are a major source of annoyance. I hit this one today, for the first time in a few years. Background info - a v210 was rather abruptly powered down and feeling somewhat ill. So I logged onto the sc and got to my console, and type boot as one does.
{1} ok boot
FATAL: system is not bootable, boot command is disabled
Which is about as helpful as someone telling me the box is currently a brick. Which I know already.
Anyway just in case you happen to hit this the fix/workaround is to set auto-boot? to false, reset the box, and then set it to true and finally boot as shown below.
{1} ok setenv auto-boot? false
auto-boot? = false
{1} ok reset-all
SC Alert: Host System has Reset
Sun Fire V210, No Keyboard
Copyright 1998-2003 Sun Microsystems, Inc. All rights reserved.
OpenBoot 4.11.4, 4096 MB memory installed, Serial #xxxxxxxx.
Ethernet address 0:3:ba:xx:xx:xx, Host ID: 83xxxxxx.
{1} ok setenv auto-boot? true
auto-boot? = true
{1} ok boot
.......... lots of output ........
volume management starting.
The system is ready.
xxxxxx console login:
(2005-04-19 20:31:07.0)
Permalink
And Nicky V. is onboard....
One of the other folks in my group, Nicky Veitch, has started his blog, joining Sean, Gleb and myself. Of interest straight away are his notes on filebench - a benchmark that will be released soon onto sourceforge ( Richard leads the design and development off filebench).
I kinda like the current parking spot for his bike as well ;).
(2005-04-19 19:55:57.0)
Permalink
Thursday April 07, 2005
Richard McDougal and Jim Mauro on bsc
I noticed that Richard McDougal has started a blog, joining his partner in crime Jim Mauro. These guys authored Solaris Internals, a book that is used by everyone in our group on an almost daily basis. They are currently updating it for Solaris 10, it will be an invaluable resource once it comes out. Point your aggregators at their blogs, highly recommended.
In the meantime they have some teasers in the form of two presentations posted on the Solaris Internals site, well worth a read.
(2005-04-07 03:45:37.0)
Permalink
Saturday March 12, 2005
Matching Funds for the Tsumani Appeal
I've avoided non technical posts for a while, but this one should be mentioned. Sun matched any donations that employees made to Tsunami Relief funds through the Sun Microsystems Foundation. I got the following mail in my inbox today (edited as the amount is my own business).
Date: March 11, 2005
Thank you for using the Sun Microsystems Foundation, Inc.
Matching Gifts Program. This is to advise you that we have
received your request to match your $personal donation to
American National Red Cross (for Tsunami Disaster Relief)
dated January 4, 2005.
Matching gift payments are made on a quarterly basis. A
matching gift payment of $personal will be sent to American
National Red Cross - Fairfax, VA, by the end of April 2005.
People have asked me why I work in Sun, theres several thousand examples of why on blogs.sun.com, as an engineer it was a place I want (and wanted in college) to work in, but its intangible things such as the reasons behind the mail above that just push it that little bit beyond other places I have worked in. (I can hear the naysayers going "oh, but everywhere does this now", well Sun's being doing it for quite a bit longer, and in many, many ways, and long before CSR became a buzzword in board rooms - with the consultants, consultancy fees and press releases that go with such buzzwords).
I mentioned this over on my personal blog as well.
(2005-03-11 18:25:54.0)
Permalink
Thursday December 09, 2004
An appserver gotcha and work around
Got stung with this for a while today, so for the benefit of others....
I had just installed an internal development version of the Sun Java System Application Server and restored a SpecJAppServer2002 domain. Within in a domain you have a script startserv (and another stopserv) which are called by the asadmin commmand when you start and stop a domain respectively.
So I've restored the domain, and go to startup my appserver
perf-n# ./asadmin start-domain -u admin spec2002-1
Starting Domain spec2002-1, please wait.
Log redirected to /var/opt/SUNWappserver/domains/spec2002-1/logs/server.log.
/var/opt/SUNWappserver/domains/spec2002-1/bin/startserv: cannot execute
CLI156 Could not start the domain spec2002-1.
But now my confusion starts
perf-n# ls -al /var/opt/SUNWappserver/domains/spec2002-1/bin/startserv
-rwxr----- 1 ecuser nobody 3152 Dec 9 02:01 /var/opt/SUNWappserver/domains/spec2002-1/bin/startserv
Looks pretty executable to me.... and I can start it manually.....
Anyway after a bit of fiddling around I figured out that the problem is that you need to have startserv with permissions of 755, a simple chmod 755 startserv and everything is back in business.
In reality your very unlikely to hit this problem if you go through the normal steps for creating and restoring a domain, but as I said on the off chance that your taking a shortcut like me.
(2004-12-08 19:55:58.0)
Permalink
|
