Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*The\s+ultimate\s+TOE\s+design\s*$/: 140 ]

Total 140 documents matching your query.

121. Re: The ultimate TOE design (score: 1)
Author: Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>
Date: Thu, 16 Sep 2004 13:57:47 +0100
You are correct - 6W average 9W TDP, still less than my scsicontroller 8)
/archives/netdev/2004-09/msg02088.html (11,755 bytes)

122. RE: The ultimate TOE design (score: 1)
Author: "Leonid Grossman" <leonid.grossman@xxxxxxxx>
Date: Thu, 16 Sep 2004 07:57:48 -0700
Well.. Any business has risks, this one doesn't seem to be higher than others :-) I view 18-26 mo timeframe as a start of the offload mass-adoption, not the end of it. In our tests, the bus contenti
/archives/netdev/2004-09/msg02090.html (16,845 bytes)

123. Re: The ultimate TOE design (score: 1)
Author: Nivedita Singhvi <niv@xxxxxxxxxx>
Date: Thu, 16 Sep 2004 09:18:49 -0700
We can dream about benefits of huge MTUs, but the reality is that moving beyond 9k MTU is years away. Reasons - mainly infrastructure, plus MTU above ~10k may loose checksum protection (granted, thi
/archives/netdev/2004-09/msg02093.html (10,232 bytes)

124. RE: The ultimate TOE design (score: 1)
Author: "Leonid Grossman" <leonid.grossman@xxxxxxxx>
Date: Thu, 16 Sep 2004 13:34:44 -0700
It's a bit painful to setup, but in general with 9k jumbos and TSO we were able to get close to pci-x 133 limit - both in LAN and WAN tests. Leonid
/archives/netdev/2004-09/msg02107.html (10,542 bytes)

125. Re: The ultimate TOE design (score: 1)
Author: Lincoln Dale <ltd@xxxxxxxxx>
Date: Fri, 17 Sep 2004 08:37:17 +1000
Hi Alan, You are correct - 6W average 9W TDP, still less than my scsicontroller 8) sure -- ok -- that gets you the main processor. now add to that a Northbridge (perhaps AMD doesnt need that but i'm
/archives/netdev/2004-09/msg02112.html (13,156 bytes)

126. Re: The ultimate TOE design (score: 1)
Author: Eric Mudama <edmudama@xxxxxxxxx>
Date: Fri, 17 Sep 2004 00:46:59 -0600
Disks have had file systems on them since close to the beginning...
/archives/netdev/2004-09/msg02132.html (10,122 bytes)

127. Re: The ultimate TOE design (score: 1)
Author: Jörn Engel <joern@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 17 Sep 2004 15:38:09 +0200
FWIW, I've already been working with complete systems that suck their power from PCI. They do exist, just not in the grocery store next door. Jörn -- Das Aufregende am Schreiben ist es, eine Ordnung
/archives/netdev/2004-09/msg02162.html (12,922 bytes)

128. Re: The ultimate TOE design (score: 1)
Author: Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>
Date: Fri, 17 Sep 2004 15:15:28 +0100
This is essentially the path Lustre is taking. Although it seems you don't want to have a "full" file system on the disk since you lose to much flexibility, instead you want the ability to allocate b
/archives/netdev/2004-09/msg02167.html (10,775 bytes)

129. Re: The ultimate TOE design (score: 1)
Author: Valdis.Kletnieks@xxxxxx
Date: Fri, 17 Sep 2004 16:27:31 -0400
e: No, he means "offload the processing of the filesystem to the disk itself". IBM's MVS systems basically did that - it used the disk's "Search Key" I/O opcodes to basically get the equivalent of do
/archives/netdev/2004-09/msg02188.html (11,161 bytes)

130. Re: The ultimate TOE design (score: 1)
Author: David Lang <david.lang@xxxxxxxxxxxxxxxxxx>
Date: Fri, 17 Sep 2004 13:36:14 -0700 (PDT)
actually the sector based access that is made to modern drives is a very primitive filesystem. if you go back to the days of the MFM and RLL drives you had the computer sending the raw bitstreams to
/archives/netdev/2004-09/msg02189.html (12,870 bytes)

131. Re: The ultimate TOE design (score: 1)
Author: Tony Lee <tony.p.lee@xxxxxxxxx>
Date: Fri, 17 Sep 2004 16:20:53 -0700
Maybe next evolutionary step is to put VFS layer directory on top of RDMA -> PCI Express/Latest serial IO, etc. Similar to access file thru NFS/SMB just on a faster standardize (RDMA) transport. On t
/archives/netdev/2004-09/msg02193.html (11,982 bytes)

132. RE: The ultimate TOE design (score: 1)
Author: "Leonid Grossman" <leonid.grossman@xxxxxxxx>
Date: Fri, 17 Sep 2004 16:36:12 -0700
Skipped... Hi Tony, For 10GbE card, we designed our own ASIC - embedded CPUs don't cut it at 10GbE... Leonid
/archives/netdev/2004-09/msg02194.html (10,449 bytes)

133. Re: The ultimate TOE design (score: 1)
Author: Nivedita Singhvi <niv@xxxxxxxxxx>
Date: Wed, 22 Sep 2004 13:18:07 -0700
From: Nivedita Singhvi [mailto:niv@xxxxxxxxxx] Sent: Thursday, September 16, 2004 9:19 AM To: Leonid Grossman Cc: 'Andi Kleen'; 'David S. Miller'; 'John Heffner'; netdev@xxxxxxxxxxx Subject: Re: The
/archives/netdev/2004-09/msg02468.html (11,630 bytes)

134. Re: The ultimate TOE design (score: 1)
Author: Eric Mudama <edmudama@xxxxxxxxx>
Date: Wed, 22 Sep 2004 17:25:38 -0600
I know what was meant. I'm not saying the filesystem on the drive is very advanced, but it's still a filesystem. Our "Record ID" is the LBA identifier, and all records are 1 block in size. We can han
/archives/netdev/2004-09/msg02489.html (10,846 bytes)

135. RE: The ultimate TOE design (score: 1)
Author: "Leonid Grossman" <leonid.grossman@xxxxxxxx>
Date: Wed, 22 Sep 2004 21:46:47 -0700
Define specific environment :-). We are running common tcp benchmarks like nttcp or iperf or Chariot or filesystem applications on a very generic white boxes, with generic OS/settings. This is somew
/archives/netdev/2004-09/msg02495.html (10,810 bytes)

136. Re: The ultimate TOE design (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Fri, 24 Sep 2004 15:07:38 +0200
Intel's IXP2800 can do 10GbE. http://www.intel.com/design/network/products/npfamily/ixp2800.htm --L
/archives/netdev/2004-09/msg02582.html (10,371 bytes)

137. Re: The ultimate TOE design (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Fri, 24 Sep 2004 15:11:58 +0200
I'm working on open source microengine code for the IXP line, which should be available Real Soon Now(TM). --L
/archives/netdev/2004-09/msg02583.html (10,484 bytes)

138. RE: The ultimate TOE design (score: 1)
Author: "Leonid Grossman" <leonid.grossman@xxxxxxxx>
Date: Fri, 24 Sep 2004 06:21:35 -0700
Hi Lennert, I was referring to the server side. One can certanly build a 10GbE box based on IPX2800 (or some other parts), but at 17-25W it is not usable in NICs since the entire PCI card budget is
/archives/netdev/2004-09/msg02584.html (10,948 bytes)

139. Re: The ultimate TOE design (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Fri, 24 Sep 2004 20:09:15 +0200
Hello, Ah, ok, that makes sense. As someone else also noted, the IXP2800 only has a 64/66 PCI interface anyway, so it wouldn't really be suitable for the task you were referring to. cheers, Lennert
/archives/netdev/2004-09/msg02591.html (10,791 bytes)

140. Re: The ultimate TOE design (score: 1)
Author: Joel Jaeggli <joelja@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 24 Sep 2004 12:39:01 -0700 (PDT)
I was referring to the server side. One can certanly build a 10GbE box based on IPX2800 (or some other parts), but at 17-25W it is not usable in NICs since the entire PCI card budget is less than th
/archives/netdev/2004-09/msg02594.html (12,437 bytes)


This search system is powered by Namazu