By: Joern Engel (joern.delete@this.logfs.org), January 7, 2009 12:52 am
Room: Moderated Discussions
anon (anon@anon.com) on 1/6/09 wrote:
---------------------------
>Joern Engel (joern@logfs.org) on 1/6/09 wrote:
>>
>>Except that certain system software defines constants like L1_CACHE_BYTES. I
>>wonder what those are all about. ;)
>
>This is a totally different issue. This is a property of the host CPU which the
>software is running on (the embedded software in the flash controller
>could define a similar constant).
Oh, yes, you are right of course. It is not a property of DRAM, but
one of the DRAM interface. And said interface has migrated into the
CPU, it is no longer part of the chipset. Don't hold it against me,
but that does look like an interface that has changed over the
years.
Software had a fairly small part in the change. It only had to learn
about cache line sizes and alignment. The bulk of it (and I agree it
is a lot) has been between DRAM, CPU and the now-gone pieces of
chipset.
>If you can propose a way to hide the details of cacheline size from software in
>a way that results in a better end result, we're all ears. Byte-sized cachelines? ;)
Well, contrary to my predictions, cache line sizes have actually
shrunk over the last generation, going back from 128B to 64B. Took me
by surprise and I am delighted. However, I see fundamental technical
reasons why cachelines will always exist and always be much bigger
than a single byte. So as you noticed, we can stand on our heads and
wiggle our toes all day, software will run faster if we use
L1_CACHE_BYTES and align our data.
Going back to the main subject, the interface between flash and the
CPU, I likewise claim that things will run faster if we know erase
sizes and align data. And more robustly if we do wear leveling
ourselves.
There is one decisive difference between DRAM and flash, though. If
the market goes along with Linus, there will effectively be a
specialized filesystem on each SSD. And things no longer run faster
when properly aligned _on top of that filesystem_. They would only
run faster if the filesystem could be bypassed somehow. DRAM is just
too low-level for anyone to put a filesystem between it and the CPU.
---------------------------
>Joern Engel (joern@logfs.org) on 1/6/09 wrote:
>>
>>Except that certain system software defines constants like L1_CACHE_BYTES. I
>>wonder what those are all about. ;)
>
>This is a totally different issue. This is a property of the host CPU which the
>software is running on (the embedded software in the flash controller
>could define a similar constant).
Oh, yes, you are right of course. It is not a property of DRAM, but
one of the DRAM interface. And said interface has migrated into the
CPU, it is no longer part of the chipset. Don't hold it against me,
but that does look like an interface that has changed over the
years.
Software had a fairly small part in the change. It only had to learn
about cache line sizes and alignment. The bulk of it (and I agree it
is a lot) has been between DRAM, CPU and the now-gone pieces of
chipset.
>If you can propose a way to hide the details of cacheline size from software in
>a way that results in a better end result, we're all ears. Byte-sized cachelines? ;)
Well, contrary to my predictions, cache line sizes have actually
shrunk over the last generation, going back from 128B to 64B. Took me
by surprise and I am delighted. However, I see fundamental technical
reasons why cachelines will always exist and always be much bigger
than a single byte. So as you noticed, we can stand on our heads and
wiggle our toes all day, software will run faster if we use
L1_CACHE_BYTES and align our data.
Going back to the main subject, the interface between flash and the
CPU, I likewise claim that things will run faster if we know erase
sizes and align data. And more robustly if we do wear leveling
ourselves.
There is one decisive difference between DRAM and flash, though. If
the market goes along with Linus, there will effectively be a
specialized filesystem on each SSD. And things no longer run faster
when properly aligned _on top of that filesystem_. They would only
run faster if the filesystem could be bypassed somehow. DRAM is just
too low-level for anyone to put a filesystem between it and the CPU.
Topic | Posted By | Date |
---|---|---|
First Dunnington benchmark results | Michael S | 2008/08/19 10:54 AM |
First Dunnington benchmark results | rwessel | 2008/08/19 01:42 PM |
First Dunnington benchmark results | Aaron Apink | 2008/08/19 05:49 PM |
First Dunnington benchmark results | Joe Chang | 2008/08/19 06:28 PM |
First Dunnington benchmark results | rwessel | 2008/08/21 09:49 AM |
First Dunnington benchmark results | Joe Chang | 2008/08/21 03:10 PM |
First Dunnington benchmark results | rwessel | 2008/08/21 06:42 PM |
First Dunnington benchmark results | Joe Chang | 2008/08/21 07:12 PM |
First Dunnington benchmark results | rwessel | 2008/08/21 09:45 AM |
First Dunnington benchmark results | Aaron Spink | 2008/08/21 01:12 PM |
First Dunnington benchmark results | Joe Chang | 2008/08/21 03:15 PM |
First Dunnington benchmark results | Richard Cownie | 2008/08/20 02:59 AM |
First Dunnington benchmark results | Anders Jensen | 2008/08/20 03:26 AM |
+SSD | Anders Jensen | 2008/08/20 03:30 AM |
First Dunnington benchmark results | Richard Cownie | 2008/08/20 11:04 AM |
First Dunnington benchmark results | slacker | 2008/08/20 12:35 PM |
First Dunnington benchmark results | Doug Siebert | 2008/08/20 07:54 PM |
First Dunnington benchmark results | Richard Cownie | 2008/08/20 08:58 PM |
SLC vs. MLC | David Kanter | 2008/08/21 01:16 AM |
SLC vs. MLC | Matt Sayler | 2008/08/21 06:25 AM |
SLC vs. MLC | Richard Cownie | 2008/08/21 06:32 AM |
SLC vs. MLC | Linus Torvalds | 2008/08/21 08:39 AM |
SLC vs. MLC | Michael S | 2008/08/21 09:07 AM |
SLC vs. MLC | Linus Torvalds | 2008/08/21 09:52 AM |
SLC vs. MLC | Michael S | 2008/08/21 10:35 AM |
OLTP appliance = mainframe? (NT) | Potatoswatter | 2008/08/21 11:44 AM |
OLTP appliance = HP NonStop? | Michael S | 2008/08/21 12:03 PM |
OLTP appliance | Joe Chang | 2008/08/21 03:33 PM |
OLTP appliance | Potatoswatter | 2008/08/21 03:59 PM |
SLC vs. MLC | Aaron Spink | 2008/08/21 01:29 PM |
SLC vs. MLC | Dan Downs | 2008/08/21 11:33 AM |
SLC vs. MLC | rwessel | 2008/08/21 12:45 PM |
SLC vs. MLC | Dan Downs | 2008/08/22 08:21 AM |
SLC vs. MLC | Aaron Spink | 2008/08/21 01:34 PM |
SLC vs. MLC vs DRAM | pgerassi | 2008/08/21 12:24 PM |
SLC vs. MLC vs DRAM | David Kanter | 2008/08/22 01:31 AM |
SLC vs. MLC | Groo | 2008/08/23 12:52 PM |
SLC vs. MLC | Doug Siebert | 2008/08/21 06:14 PM |
SLC vs. MLC | Linus Torvalds | 2008/08/22 08:05 AM |
SLC vs. MLC | Doug Siebert | 2008/08/22 02:27 PM |
SLC vs. MLC | EduardoS | 2008/08/22 06:26 PM |
SSD Controller differentiation | David Kanter | 2008/08/22 09:35 PM |
SSD Controller differentiation | Doug Siebert | 2008/08/22 10:34 PM |
SSD Controller differentiation (supercaps, cost...) | anon | 2008/08/23 10:18 AM |
SSD Controller differentiation (supercaps, cost...) | Doug Siebert | 2008/08/23 10:40 AM |
SLC vs. MLC | Linus Torvalds | 2008/08/23 10:50 AM |
SLC vs. MLC | Linus Torvalds | 2008/09/08 12:03 PM |
SLC vs. MLC | Max | 2008/09/08 01:51 PM |
SLC vs. MLC | Howard Chu | 2008/09/08 09:04 PM |
SLC vs. MLC | Max | 2008/09/08 10:29 PM |
SLC vs. MLC | Howard Chu | 2008/09/09 12:12 AM |
RAM vs SSD? | Jouni Osmala | 2008/09/09 01:06 AM |
RAM vs SSD? | Max | 2008/09/12 12:51 PM |
RAM vs SSD? | EduardoS | 2008/09/12 04:27 PM |
Disk cache snapshotting | Max | 2008/09/13 08:34 AM |
Disk cache snapshotting | Howard Chu | 2008/09/14 09:58 PM |
Disk cache snapshotting | Max | 2008/09/15 12:50 PM |
SLC vs. MLC | Linus Torvalds | 2008/09/09 07:43 AM |
SLC vs. MLC | Howard Chu | 2008/09/09 09:42 AM |
SLC vs. MLC | Linus Torvalds | 2008/09/09 10:39 AM |
SLC vs. MLC | Michael S | 2008/09/10 12:29 AM |
SLC vs. MLC | anon | 2008/09/10 02:51 AM |
SLC vs. MLC | Michael S | 2008/09/10 03:09 AM |
SLC vs. MLC | Max | 2008/09/10 04:48 AM |
SLC vs. MLC | Michael S | 2008/09/10 05:52 AM |
SLC vs. MLC | Max | 2008/09/10 06:28 AM |
SLC vs. MLC | Matt Sayler | 2008/09/10 06:21 AM |
SLC vs. MLC | Michael S | 2008/09/10 09:17 AM |
SLC vs. MLC | anon | 2008/09/10 06:29 AM |
SLC vs. MLC | Michael S | 2008/09/10 09:23 AM |
SLC vs. MLC | Matt Sayler | 2008/09/10 10:45 AM |
SLC vs. MLC | Linus Torvalds | 2008/09/10 07:25 AM |
SLC vs. MLC | Michael S | 2008/09/10 09:54 AM |
SLC vs. MLC | Linus Torvalds | 2008/09/10 10:31 AM |
Physical vs effective write latency | Max | 2008/09/11 07:35 AM |
Physical vs effective write latency | Linus Torvalds | 2008/09/11 09:06 AM |
Physical vs effective write latency | Linus Torvalds | 2008/09/11 09:48 AM |
Physical vs effective write latency | Linus Torvalds | 2008/09/11 11:39 AM |
Physical vs effective write latency | Mark Roulo | 2008/09/11 12:18 PM |
Physical vs effective write latency | Doug Siebert | 2008/09/11 05:59 PM |
Physical vs effective write latency | Linus Torvalds | 2008/09/11 07:16 PM |
Physical vs effective write latency | Doug Siebert | 2008/09/11 10:28 PM |
Physical vs effective write latency | MS | 2009/02/03 03:06 PM |
SLC vs. MLC - the trick to latency | Anonymous | 2008/09/11 12:39 PM |
SLC vs. MLC - the trick to latency | anon | 2008/09/11 01:17 PM |
SLC vs. MLC - the trick to latency | Anonymous | 2008/09/11 05:25 PM |
SLC vs. MLC - the trick to latency | Doug Siebert | 2008/09/11 05:47 PM |
SLC vs. MLC - the trick to latency | rwessel | 2008/09/11 06:01 PM |
SLC vs. MLC - the trick to latency | anon | 2008/09/12 12:00 AM |
SLC vs. MLC - the trick to latency | Anonymous | 2008/09/12 08:52 PM |
SLC vs. MLC - the trick to latency | anon | 2008/09/13 10:06 AM |
SLC vs. MLC - the trick to latency | Ungo | 2008/09/15 12:18 PM |
To SSD or not? One lady's perspective | David Kanter | 2008/09/22 01:12 AM |
To SSD or not? One lady's perspective | Howard Chu | 2008/09/22 04:02 AM |
To SSD or not? Real data.. | Linus Torvalds | 2008/09/22 07:33 AM |
To SSD or not? Real data.. | Ungo | 2008/09/22 12:27 PM |
4K sectors | Wes Felter | 2008/09/22 06:03 PM |
4K sectors | Daniel | 2008/09/22 10:31 PM |
Reasons for >512 byte sectors | Doug Siebert | 2008/09/22 09:38 PM |
Reasons for >512 byte sectors | rwessel | 2008/09/22 10:09 PM |
Reasons for >512 byte sectors | Howard Chu | 2008/09/23 02:50 AM |
Reasons for >512 byte sectors | Daniel | 2008/09/22 10:40 PM |
Reasons for >512 byte sectors | rwessel | 2008/09/23 09:11 AM |
Reasons for >512 byte sectors | Daniel | 2008/09/23 12:10 PM |
HDD long sector size availability | Etienne Lehnart | 2008/09/23 05:32 AM |
HDD long sector size availability | rwessel | 2008/09/23 09:19 AM |
HDD long sector size availability | Etienne Lehnart | 2008/09/23 02:17 PM |
To SSD or not? Real data.. | Jouni Osmala | 2008/09/22 11:16 PM |
To SSD or not? One lady's perspective | Wes Felter | 2008/09/22 11:25 AM |
How should SSDs be engineered into systems? | Rob Thorpe | 2008/09/22 02:01 PM |
How should SSDs be engineered into systems? | Matt Craighead | 2008/09/23 06:59 PM |
How should SSDs be engineered into systems? | Matt Sayler | 2008/09/24 04:17 AM |
ATA/SCSIS, Write Flushes and Asych Filesystems | TruePath | 2009/01/25 04:44 AM |
SLC vs. MLC - the trick to latency | Michael S | 2008/09/12 04:58 AM |
overlapped erase and read | Michael S | 2008/09/12 04:59 AM |
overlapped erase and read | David W. Hess | 2008/09/12 09:56 AM |
overlapped erase and read | Anonymous | 2008/09/12 08:45 PM |
overlapped erase and read | Jouni Osmala | 2008/09/12 11:56 PM |
overlapped erase and read | Michael S | 2008/09/13 11:29 AM |
overlapped erase and read | Michael S | 2008/09/13 12:09 PM |
overlapped erase and read | Linus Torvalds | 2008/09/13 02:05 PM |
SLC vs. MLC - the trick to latency | Doug Siebert | 2008/09/11 05:31 PM |
SLC vs. MLC | EduardoS | 2008/09/08 02:07 PM |
SLC vs. MLC | Linus Torvalds | 2008/09/08 02:30 PM |
SLC vs. MLC | EduardoS | 2008/09/08 04:01 PM |
SSD and RAID | Joe Chang | 2008/09/08 07:42 PM |
SSD and RAID | Doug Siebert | 2008/09/08 09:46 PM |
SSD and RAID | Aaron Spink | 2008/09/09 04:27 PM |
SSD and RAID | Groo | 2008/09/10 01:02 PM |
SLC vs. MLC | Joern Engel | 2009/01/06 10:22 AM |
SLC vs. MLC | Linus Torvalds | 2009/01/06 02:04 PM |
SLC vs. MLC | Joern Engel | 2009/01/06 03:24 PM |
SLC vs. MLC | rwessel | 2009/01/06 04:47 PM |
SLC vs. MLC | anonymous | 2009/01/06 05:17 PM |
SLC vs. MLC | rwessel | 2009/01/06 05:58 PM |
SLC vs. MLC | Joern Engel | 2009/01/07 12:35 AM |
SLC vs. MLC | Linus Torvalds | 2009/01/06 05:45 PM |
SLC vs. MLC | rwessel | 2009/01/06 06:09 PM |
SLC vs. MLC | Linus Torvalds | 2009/01/06 07:47 PM |
SLC vs. MLC | Joern Engel | 2009/01/07 12:26 AM |
SLC vs. MLC | anon | 2009/01/06 08:23 PM |
SLC vs. MLC | Joern Engel | 2009/01/07 12:52 AM |
SLC vs. MLC | anon | 2009/01/07 02:34 AM |
SLC vs. MLC | IntelUser2000 | 2009/01/07 07:43 AM |
SLC vs. MLC | Linus Torvalds | 2009/01/07 10:28 AM |
drop data filesystem semantic | Doug Siebert | 2009/01/09 12:21 PM |
FTL and FS | iz | 2009/01/09 07:49 PM |
FTL and FS | Linus Torvalds | 2009/01/09 09:53 PM |
FTL and FS | iz | 2009/01/10 02:09 AM |
FTL and FS | Michael S | 2009/01/10 03:19 PM |
compiling large programs | iz | 2009/01/10 05:51 PM |
compiling large programs | Linus Torvalds | 2009/01/10 07:58 PM |
compiling large programs | peter | 2009/01/11 05:30 AM |
compiling large programs | Andi Kleen | 2009/01/11 01:03 PM |
The File Abstraction | TruePath | 2009/01/25 06:45 AM |
The File Abstraction | Howard Chu | 2009/01/25 01:49 PM |
The File Abstraction | Linus Torvalds | 2009/01/26 09:23 AM |
The File Abstraction | Michael S | 2009/01/26 01:39 PM |
The File Abstraction | Linus Torvalds | 2009/01/26 02:31 PM |
The File Abstraction | Dean Kent | 2009/01/26 03:06 PM |
The File Abstraction | Linus Torvalds | 2009/01/26 04:29 PM |
The File Abstraction | Mark Christiansen | 2009/01/27 09:24 AM |
The File Abstraction | Mark Christiansen | 2009/01/27 10:14 AM |
The File Abstraction | Linus Torvalds | 2009/01/27 10:15 AM |
The File Abstraction | slacker | 2009/01/27 11:20 AM |
The File Abstraction | Linus Torvalds | 2009/01/27 01:16 PM |
Attributes All The Way Down | Mark Christiansen | 2009/01/27 02:17 PM |
The File Abstraction | slacker | 2009/01/27 05:25 PM |
The File Abstraction | Linus Torvalds | 2009/01/28 08:17 AM |
The File Abstraction: API thoughts | Carlie Coats | 2009/01/28 09:35 AM |
The File Abstraction | slacker | 2009/01/28 10:09 AM |
The File Abstraction | Linus Torvalds | 2009/01/28 01:44 PM |
Programs already 'hide' their metadata in the bytestream, unbeknownst to users | anon | 2009/01/28 09:28 PM |
The File Abstraction | slacker | 2009/01/29 10:39 AM |
The File Abstraction | Linus Torvalds | 2009/01/29 11:08 AM |
The File Abstraction | Dean Kent | 2009/01/29 11:49 AM |
The File Abstraction | Howard Chu | 2009/01/29 02:58 PM |
The File Abstraction | rwessel | 2009/01/29 04:23 PM |
Extended Attributes in Action | slacker | 2009/01/29 03:05 PM |
Extended Attributes in Action | stubar | 2009/01/29 04:49 PM |
Extended Attributes in Action | Linus Torvalds | 2009/01/29 05:15 PM |
Like Duh | anon | 2009/01/29 07:42 PM |
Like Duh | anon | 2009/01/29 09:15 PM |
Like Duh | anon | 2009/02/01 07:18 PM |
Double Duh. | Anonymous | 2009/02/01 10:58 PM |
Double Duh. | anon | 2009/02/02 02:08 AM |
Double Duh. | Anonymous | 2009/02/02 05:11 PM |
Double Duh. | anon | 2009/02/02 07:33 PM |
Like Duh | David Kanter | 2009/02/01 11:05 PM |
Like Duh | peter | 2009/02/01 11:55 PM |
Like Duh | anon | 2009/02/02 01:55 AM |
Xattrs, Solar power, regulation and politics | Rob Thorpe | 2009/02/02 04:36 AM |
Terminology seems too fuzzy to me | hobold | 2009/02/02 06:14 AM |
Terminology seems too fuzzy to me | rwessel | 2009/02/02 12:33 PM |
good summary | Michael S | 2009/02/03 02:41 AM |
good summary | Mark Christiansen | 2009/02/03 09:57 AM |
good summary | Howard Chu | 2009/02/03 10:21 AM |
good summary | Mark Christiansen | 2009/02/03 11:18 AM |
good summary | Howard Chu | 2009/02/03 12:00 PM |
good summary | Mark Christiansen | 2009/02/03 12:36 PM |
good summary | RagingDragon | 2009/02/03 10:39 PM |
good summary | rwessel | 2009/02/03 11:03 PM |
good summary | RagingDragon | 2009/02/03 11:46 PM |
Terminology seems too fuzzy to me | slacker | 2009/02/04 05:06 PM |
Terminology seems too fuzzy to me | Michael S | 2009/02/05 01:05 AM |
Terminology seems too fuzzy to me | Ungo | 2009/02/05 01:15 PM |
Terminology seems too fuzzy to me | slacker | 2009/02/05 02:19 PM |
Terminology seems too fuzzy to me | Howard Chu | 2009/02/05 04:44 PM |
Like Duh | iz | 2009/01/30 02:03 AM |
EAs (security labels) hosed me badly | anon | 2009/01/30 09:48 PM |
Extended Attributes in Action | RagingDragon | 2009/01/29 09:31 PM |
Extended Attributes in Action | anonymous | 2009/01/29 08:13 PM |
Extended Attributes in Action | Howard Chu | 2009/01/29 09:38 PM |
Extended Attributes in Action | slacker | 2009/01/30 11:24 AM |
Extended Attributes in Action | anon | 2009/01/30 05:50 PM |
Extended Attributes in Action | Etienne Lehnart | 2009/01/30 12:22 AM |
Extended Attributes in Action | Rob Thorpe | 2009/01/30 12:39 PM |
Extended Attributes in Action | slacker | 2009/01/30 01:16 PM |
Extended Attributes in Action | anon | 2009/01/30 06:03 PM |
Extended Attributes in Action | Howard Chu | 2009/01/30 11:22 PM |
Extended Attributes in Action | rwessel | 2009/01/31 12:08 AM |
Extended Attributes in Action | anonymous | 2009/01/31 12:22 AM |
Extended Attributes in Action | rwessel | 2009/01/31 12:56 AM |
Scaling | Dean Kent | 2009/01/31 09:04 AM |
Scaling | Rob Thorpe | 2009/02/02 02:39 AM |
Scaling | rwessel | 2009/02/02 11:41 AM |
Scaling | Howard Chu | 2009/02/02 12:30 PM |
Scaling | Dean Kent | 2009/02/02 02:27 PM |
Scaling | Rob Thorpe | 2009/02/03 05:08 AM |
Scaling | Dean Kent | 2009/02/03 07:38 AM |
Scaling | rwessel | 2009/02/03 02:34 PM |
Scaling | RagingDragon | 2009/02/03 10:46 PM |
in defense of software that does not scale | Matt Sayler | 2009/02/03 11:27 AM |
in defense of software that does not scale | Howard Chu | 2009/02/03 12:03 PM |
in defense of software that does not scale | Matt Sayler | 2009/02/03 12:17 PM |
in defense of software that does not scale | RagingDragon | 2009/02/03 11:00 PM |
in defense of software that does not scale | Michael S | 2009/02/04 06:46 AM |
in defense of software that does not scale | RagingDragon | 2009/02/04 09:33 PM |
in defense of software that does not scale | Dean Kent | 2009/02/03 12:17 PM |
in defense of software that does not scale | Matt Sayler | 2009/02/03 12:24 PM |
in defense of software that does not scale | Vincent Diepeveen | 2009/02/04 10:43 AM |
in defense of software that does not scale | rwessel | 2009/02/03 02:44 PM |
in defense of software that does not scale | anon | 2009/02/04 02:35 AM |
in defense of software that does not scale | Carlie Coats | 2009/02/04 05:24 AM |
Scaling with time vs. scaling from the beginning. | mpx | 2009/02/05 01:57 AM |
Extended Attributes in Action | Michael S | 2009/01/31 10:33 AM |
Extended Attributes in Action | anon | 2009/01/31 10:37 PM |
Extended Attributes in Action | JasonB | 2009/01/31 08:11 AM |
Extended Attributes in Action | Howard Chu | 2009/01/31 11:43 AM |
Extended Attributes in Action | JasonB | 2009/01/31 04:37 PM |
Extended Attributes in Action | Howard Chu | 2009/02/02 02:42 PM |
Extended Attributes in Action | Howard Chu | 2009/02/02 02:44 PM |
The File Abstraction | Rob Thorpe | 2009/01/27 11:20 AM |
The File Abstraction | Howard Chu | 2009/01/27 12:28 AM |
The File Abstraction | Michael S | 2009/01/27 03:00 AM |
The File Abstraction | Dean Kent | 2009/01/27 08:30 AM |
The File Abstraction | Andi Kleen | 2009/01/27 02:05 AM |
SLC vs. MLC | Michel | 2009/01/12 06:54 PM |
SLC vs. MLC | Linus Torvalds | 2009/01/12 07:38 PM |
SLC vs. MLC | rwessel | 2009/01/13 12:52 AM |
SLC vs. MLC | Ungo | 2009/01/13 03:04 PM |
SLC vs. MLC | Wes Felter | 2009/01/13 05:42 PM |
SLC vs. MLC | TruePath | 2009/01/25 05:05 AM |
SLC vs. MLC | Ungo | 2008/08/21 12:54 PM |
SLC vs. MLC | Aaron Spink | 2008/08/21 01:20 PM |
MLC vs. SLC | Michael S | 2008/08/21 08:57 AM |
First Dunnington benchmark results | rwessel | 2008/08/21 10:40 AM |
First Dunnington benchmark results | Aaron Spink | 2008/08/21 03:18 AM |
First Dunnington benchmark results | Etienne Lehnart | 2008/08/20 04:38 AM |
Will x86 dominate big iron? | Tom W | 2008/08/19 10:10 PM |
Will x86 dominate big iron? | Jesper Frimann | 2008/08/20 12:28 AM |
Will x86 dominate big iron? | Tom W | 2008/08/20 03:42 PM |
Will x86 dominate big iron? | David Kanter | 2008/08/21 01:13 AM |
Will x86 dominate big iron? | Joe Chang | 2008/08/21 06:54 AM |
Will x86 dominate big iron? | asdf | 2008/08/22 01:18 PM |
Will x86 dominate big iron? | Dean Kent | 2008/08/22 07:54 PM |
Will x86 dominate big iron? | Jesper Frimann | 2008/08/22 09:48 AM |
Will x86 dominate big iron? | Tom W | 2008/08/24 01:06 AM |
Will x86 dominate big iron? | Michael S | 2008/08/24 04:19 AM |
Will x86 dominate big iron? | Dean Kent | 2008/08/24 09:30 AM |
Will x86 dominate big iron? | Paul | 2008/08/24 11:16 AM |
Will x86 dominate big iron? | Dean Kent | 2008/08/24 12:37 PM |
Will x86 dominate big iron? | Michael S | 2008/08/25 12:53 AM |
Will x86 dominate big iron? | someone | 2008/08/22 10:19 AM |
Will x86 dominate big iron? | aaron spink | 2008/08/23 02:56 AM |
Will x86 dominate big iron? | Michael S | 2008/08/23 09:58 AM |
Will x86 dominate big iron? | someone | 2008/08/23 01:51 PM |
Will x86 dominate big iron? | someone | 2008/08/23 01:55 PM |
Will x86 dominate big iron? | Aaron Spink | 2008/08/23 04:52 PM |
Will x86 dominate big iron? | anonymous | 2008/08/23 05:28 PM |
Will x86 dominate big iron? | Dean Kent | 2008/08/23 06:12 PM |
Off road and topic | EduardoS | 2008/08/23 06:28 PM |
Will x86 dominate big iron? | someone | 2008/08/23 06:26 PM |
Will x86 dominate big iron? | Dean Kent | 2008/08/23 09:40 PM |
Will x86 dominate big iron? | anonymous | 2008/08/24 01:46 AM |
Off road and topic | David W. Hess | 2008/08/24 03:24 AM |
Off road and topic | Aaron Spink | 2008/08/24 04:14 AM |
Beckton vs. Dunnington | Mr. Camel | 2008/08/22 06:30 AM |
Beckton vs. Dunnington | jokerman | 2008/08/22 12:12 PM |
Beckton vs. Dunnington | Mr. Camel | 2009/05/29 10:16 AM |