Re: 923 Mbps across the Ocean ...
On 3/7/2003 at 14:57:22 -0500, Eric Germann said:
http://www.cnn.com/2003/TECH/internet/03/07/speed.record/index.html
Comments folks?
Yeah. Give me a million dollars, plus fiber from here to anywhere, and let me muck with the TCP algorithm, and I can move a gig-e worth of traffic, too. -Dave
What am I missing here, OC48=2.5Gb, OC192=10Gb, theres native 1GigE, native 10GigE So whats good about this? On Fri, 7 Mar 2003, Dave Israel wrote:
On 3/7/2003 at 14:57:22 -0500, Eric Germann said:
http://www.cnn.com/2003/TECH/internet/03/07/speed.record/index.html
Comments folks?
Yeah. Give me a million dollars, plus fiber from here to anywhere, and let me muck with the TCP algorithm, and I can move a gig-e worth of traffic, too.
-Dave
More detailed technical information on the periodic I2 Land Speed Record contest can be found at http://lsr.internet2.edu/ The answer to "what's good about this" is left as an exercise to the reader. ---Rob "Stephen J. Wilcox" <steve@telecomplete.co.uk> writes:
What am I missing here, OC48=2.5Gb, OC192=10Gb, theres native 1GigE, native 10GigE
So whats good about this?
On Fri, 7 Mar 2003, Dave Israel wrote:
On 3/7/2003 at 14:57:22 -0500, Eric Germann said:
http://www.cnn.com/2003/TECH/internet/03/07/speed.record/index.html
Comments folks?
Yeah. Give me a million dollars, plus fiber from here to anywhere, and let me muck with the TCP algorithm, and I can move a gig-e worth of traffic, too.
-Dave
On Fri, Mar 07, 2003 at 03:43:28PM -0500, Robert E. Seastrom wrote:
More detailed technical information on the periodic I2 Land Speed Record contest can be found at http://lsr.internet2.edu/
The answer to "what's good about this" is left as an exercise to the reader.
This might have something to do with it: Contest Rules: 1. A minimum of 100 megabytes must be transferred a minimum terrestrial distance of 100 kilometers with a minimum of two router hops in each direction between the source node and the destination node across one or more operational and production-oriented high-performance research and education networks. Examples of such networks are Abilene, ESnet, CA*net3, NREN and GEANT. Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis. -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link? Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me. -- Mikael Abrahamsson email: swmike@swm.pp.se
On Fri, Mar 07, 2003 at 10:09:51PM +0100, Mikael Abrahamsson wrote:
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
Probably not many, but it's quite possible. I've done 1 Gbps streams off a single box before, it's not much harder to take it out 150ms. Heck for their 60 second test, you could buy some GigE transit ports from someone with a STM64 across the pond and not even pay for it in 95th percentile. :)
Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me.
And you can't afford 20MB of RAM because...? -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)
On Fri, Mar 07, 2003 at 10:09:51PM +0100, Mikael Abrahamsson quacked:
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me.
It's unusual, but it's not completely unheard of. One of the biggest sources of such data is VLBI (interferometry to measure the movement of the earth's crust), in which signals from geographically distributed measurement sites have to be recorded and correlated at a central site: http://web.haystack.edu/vlbi/vlbisystems.html The signals are massive. Right now they use specially made tape drives that can record 1Gb/s: ftp://web.haystack.edu/pub/mark4/memos/230.2.pdf ftp://web.haystack.edu/pub/mark4/memos/HDR_concept.PDF and they send the data around via airplanes. They'd love to be able to do real-time correlation of the data, but that involves collecting 6 of these feeds at a central site (more coming). The feeds must be capable of running unattended for up to 24 hours (86 terabytes each, or an aggregate of half a petabyte per day). Yes, backbones push more than a gigabit across links, but not as for a single flow of data. -Dave -- work: dga@lcs.mit.edu me: dga@pobox.com MIT Laboratory for Computer Science http://www.angio.net/ I do not accept unsolicited commercial email. Do not spam me.
On Friday, March 7, 2003, at 04:37 PM, David G. Andersen wrote:
On Fri, Mar 07, 2003 at 10:09:51PM +0100, Mikael Abrahamsson quacked:
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me.
It's unusual, but it's not completely unheard of. One of the biggest sources of such data is VLBI (interferometry to measure the movement of the earth's crust), in which signals from geographically distributed measurement sites have to be recorded and correlated at a central site:
http://web.haystack.edu/vlbi/vlbisystems.html
The signals are massive. Right now they use specially made tape drives that can record 1Gb/s:
ftp://web.haystack.edu/pub/mark4/memos/230.2.pdf
ftp://web.haystack.edu/pub/mark4/memos/HDR_concept.PDF
and they send the data around via airplanes. They'd love to be able to do real-time correlation of the data, but that involves collecting 6 of these feeds at a central site (more coming). The feeds must be capable of running unattended for up to 24 hours (86 terabytes each, or an aggregate of half a petabyte per day).
VLBI is moving to hard drive replacements for the expensive 1 inch tapes currently used (known as Mark V). There are active projects for "e-VLBI" - at CRL in Japan http://www.ntt.co.jp/news/news01e/0107/010706.html and at Haystack Observatory in Massachusetts http://web.haystack.edu/e-vlbi/meeting.html In e-VLBI there is no need for reliable transmission and UDP is the way to go. I am still involved with this peripherally, especially with the idea that the traffic be sent "worse than best effort", so as not to collide with regular traffic. BTW, when I did VLBI for the Navy, we used to move literally tons of tapes around the world per month and achieved sustained bandwidths > 1 Gbps, albeit with FED-EX, not routers.
Yes, backbones push more than a gigabit across links, but not as for a single flow of data.
-Dave
-- work: dga@lcs.mit.edu me: dga@pobox.com MIT Laboratory for Computer Science http://www.angio.net/ I do not accept unsolicited commercial email. Do not spam me.
Regards Marshall Eubanks T.M. Eubanks Multicast Technologies, Inc. Phone : 703-293-9601 Fax : 703-293-9609 e-mail : tme@multicasttech.com http://www.multicasttech.com Our New Multicast Workshop : http://www.multicasttech.com/workshop
Hello; There was a 1.5 Gigabit per second uncompressed HDTV stream that was sent U Washington to DC. http://www.washington.edu/hdtv/next-gen.html True, that was sent RTP over UDP, but if I was going to send that data rate over a _dedicated_ link, I would use UDP. The very high bit rate applications I have been involved with could all tolerate some loss, and for the others, a little FEC could go a long way. On Friday, March 7, 2003, at 04:09 PM, Mikael Abrahamsson wrote:
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me.
-- Mikael Abrahamsson email: swmike@swm.pp.se
Regards Marshall Eubanks T.M. Eubanks Multicast Technologies, Inc. Phone : 703-293-9601 Fax : 703-293-9609 e-mail : tme@multicasttech.com http://www.multicasttech.com Our New Multicast Workshop : http://www.multicasttech.com/workshop
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
what kind of production highway sees the kinds of cars that reach the world land speed records? why is it that nanog'ers seem to display such aggression at this type of thing? is it the article itself, the work the researchers are doing, or the fact that they had the bandwidth and hardware to do it with? --Rob
On Fri, 7 Mar 2003, fingers wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
what kind of production highway sees the kinds of cars that reach the world land speed records?
why is it that nanog'ers seem to display such aggression at this type of thing? is it the article itself, the work the researchers are doing, or the fact that they had the bandwidth and hardware to do it with?
to be fair (as I was first to flame!) it is presented out of context in a poorly written and somewhat misleading news article Steve
On Fri, Mar 07, 2003 at 11:56:29PM +0200, fingers wrote:
why is it that nanog'ers seem to display such aggression at this type of thing? is it the article itself, the work the researchers are doing, or the fact that they had the bandwidth and hardware to do it with?
There is nothing wrong with trying to set speed records, trying to push tcp performance to its limits and maybe beyond, or holding contests to do any of the above. I think the objections here are three-fold: A) The amount of arrogance it takes to declare a land speed "record" when there are people out there doing way more than this on a regular basis. B) The extreme wastefulness of spending a million dollars to do it. C) The incredible (well ok maybe not that incredible, expected is more like it) lack of accuracy in the reporting of this story. -- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)
RAS> Date: Sat, 8 Mar 2003 01:25:17 -0500 RAS> From: Richard A Steenbergen RAS> I think the objections here are three-fold: "Researchers take advantage of ideal conditions and huge funding to do a fraction of what network engineers do every day" just doesn't help ratings. This is the same mass media that predicted the end of the world when our calendars turned 2000. Alas, I suppose a thread about "American mass media stinks" would be just about as revolutionary as that on which they "reported". The difference is that NANOG posts are cheaper and at least somewhat more accurate. Eddy -- Brotsman & Dreger, Inc. - EverQuick Internet Division Bandwidth, consulting, e-commerce, hosting, and network building Phone: +1 (785) 865-5885 Lawrence and [inter]national Phone: +1 (316) 794-8922 Wichita ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Date: Mon, 21 May 2001 11:23:58 +0000 (GMT) From: A Trap <blacklist@brics.com> To: blacklist@brics.com Subject: Please ignore this portion of my mail signature. These last few lines are a trap for address-harvesting spambots. Do NOT send mail to <blacklist@brics.com>, or you are likely to be blocked.
On Sat, 8 Mar 2003, Richard A Steenbergen wrote:
A) The amount of arrogance it takes to declare a land speed "record" when there are people out there doing way more than this on a regular basis.
Single stream at 900mbs over that distance? Where? Jason -- Jason Slagle - CCNP - CCDP /"\ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . \ / ASCII Ribbon Campaign . X - NO HTML/RTF in e-mail . / \ - NO Word docs in e-mail .
On Sat, 8 Mar 2003, Richard A Steenbergen wrote:
A) The amount of arrogance it takes to declare a land speed "record" when there are people out there doing way more than this on a regular basis.
Single stream at 900mbs over that distance? Where?
Talk to folks that deal with radio telescopes. Alex
At 10:09 PM 07-03-03 +0100, Mikael Abrahamsson wrote:
On Fri, 7 Mar 2003, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
What kind of production environment needs a single TCP stream of data at 1 gigabit/s over a 150ms latency link?
Just the fact that you need a ~20 megabyte TCP window size to achieve this (feel free to correct me if I'm wrong here) seems kind of unusal to me.
See: http://www.psc.edu/networking/perf_tune.html for many details of tuning your TCP stack. Interesting that most of the links lead to places like PSC, VT, ANL, ORNL, UofHannover and NLANR. If commercial ISPs have been doing this stuff "on a regular basis", please let us know where it is documented since it is a bit well hidden. -Hank
-- Mikael Abrahamsson email: swmike@swm.pp.se
On Friday, Mar 7, 2003, at 20:53 Europe/London, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
Fortunately, these days there are very few production networks press-releasing the size of their ISPnesses. Sean. (mine's bigger than yours, anyway)
At 03:53 PM 07-03-03 -0500, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
Please document it so as to shame these I2 networks. Somehow, I doubt you will be able to. -Hank
-- Richard A Steenbergen <ras@e-gerbil.net> http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)
Thus spake "Hank Nussbacher" <hank@att.net.il>
At 03:53 PM 07-03-03 -0500, Richard A Steenbergen wrote:
Production commercial networks need not apply, 'lest someone realize that they blow away these speed records on a regular basis.
Please document it so as to shame these I2 networks. Somehow, I doubt you will be able to.
Internet/2 is not interesting because it has big pipes; the public Internet has much bigger pipes and more of them. I/2 is interesting only because it has fewer users -- by two or three orders of magnitude -- and most/all of these users are connected by FastE or better. However, there is no need to waste funding buying uber-fast routers or GigE links around the globe just to learn how to tune stacks or apps. If high-speed TCP research is what you're doing, rig up a latency generator in your laboratory and do your tests that way, just like the TCPSAT folks. Spending millions of (probably taxpayer) dollars to win a meaningless record is unethical, IMHO. S Stephen Sprunk "God does not play dice." --Albert Einstein CCIE #3723 "God is an inveterate gambler, and He throws the K5SSS dice at every possible opportunity." --Stephen Hawking
On Fri, 7 Mar 2003, Dave Israel wrote:
Yeah. Give me a million dollars, plus fiber from here to anywhere, and let me muck with the TCP algorithm, and I can move a gig-e worth of traffic, too.
Doing the 923 Mbps for one stream may be non-trivial (heck, even doing 110 MB per second sustained to/from disk isn't trivial) but the real challenge would be having two of those setups each try to do 1 Gbps over a single connecting while sharing the 1 Gbps link without any slowdowns. (And no fair having 20 MB interface buffers in the routers.)
On Fri, 7 Mar 2003, Dave Israel wrote:
On 3/7/2003 at 14:57:22 -0500, Eric Germann said:
http://www.cnn.com/2003/TECH/internet/03/07/speed.record/index.html
Comments folks?
Yeah. Give me a million dollars, plus fiber from here to anywhere, and let me muck with the TCP algorithm, and I can move a gig-e worth of traffic, too.
Or a 400,000pps attack :) Which we've seen
participants (17)
-
alex＠yuriev.com -
Christopher L. Morrow -
Dave Israel -
David G. Andersen -
E.B. Dreger -
fingers -
Hank Nussbacher -
Iljitsch van Beijnum -
Jason Slagle -
Marshall Eubanks -
Matt Zimmerman -
Mikael Abrahamsson -
Richard A Steenbergen -
rs＠seastrom.com -
Sean M.Doran -
Stephen J. Wilcox -
Stephen Sprunk