Actually many do. I'm in the media industry. Not only does my employer allow and encourage viewing listening of all manner of compressed and uncompressed content (let's not even talk about uncompressed content on 'Fister'), for many of my (000s of..) colleagues, it is a central part of their job. Next !
funnily enough, no. so why exactly does a high latency low packet loss network slow down TCP based protocols so much ? Or are you saying that this doesn't happen ? I'm genuinely interested in your answer.
.. although I keep coming back to the sentence "...senses congestion by continuously measuring the round-trip time for the TCP acknowledgment and then monitoring how that measurement changes from moment to moment.".
I would imagine in the typical high-latency scenario, where regular TCP is mis-interpreting long RTT as link congestion, and backing off the rate, FastTCP is able to actually keep pushing the rate up, meanwhile keeping an eye on the RTT. I mean, the RTT shouldn't increase in line with the rate, unless the link actually *is* congested. So just increase the rate until the RTT increases, at which point you are genuinely maxxing out the link. I think that must be how it is working..
Not so. High (>1500ms) latency *severely* affects TCP protocols like FTP. I encounter this on trans-continental WANs which go over satellite every day. I've tried some UDP compression such as FileCatalyst, and 15-20x speedup is possible on some links.
The problem is that "regular" TCP mis-interprets long Round-Trip-Time (aka latency) as link congestion and backs off the rate at which it is sending packets.
The bandwidth between point A and B may be rated at a high throughput, but TCP protocols such as FTP will never achieve that speed if the RTT is long. Increasing the bandwidth won't help !! So a slowdown of 20-30x is not uncommon on WAN links with high latency e.g. transcontinental, via satellite.
I've looked at technologies like Digital Fountain (and it's Java implementation, FileCatalyst) which use UDP and some clever mathematics to overcome latency, however it's not clear from TFA what FastTCP is doing underneath..
Hey kickdown, while I like *totally* support what you're saying about shell, the AC does link to a 'send to power toy' where you need to press Ctrl to change the directory in the dosbox from the gui. Rock on d00d.
Or you could use any of the existing videoconferencing hardware/software and actually see their real faces. Sometimes that's not what you want;o)
Sure, that's not quite as cool, Agreed, not sure how it helps you argument though.
but I bet it's several times more productive You don't offer any evidence. Do you have experience in this area ? I would say that a virtual world is more productive in some ways - for instance, in the videoconferencing scenario, there really isn't a way to continue interacting after the conference is done. Everyone disconects from the conference, and that's it. In the virtual world, it is possible to have a conversation with your virtualised colleagues as you leave the virtual meeting room i.e. 'corridor chat'.
and it already exists. metaverse platforms already exist too.
Last year I attended an excellent seminar track on content and knowledge at this : http://www.xmlsummerschool.com/ - and one of the speakers had a great example - he wanted to be able to search for a guitar amp speaker cabinet that would handle the 100w (that's RMS) output of his Marshall amp, and fit in the boot/trunk of his car - I forget the make/model, let's say it's a Ford whatever... anyhow, the point is that the semantic search app would need to discover the dimensions of the guy's car's boot/trunk from Ford, then it could search for cabs that would fit in the boot/trunk that could handle 100w (that's RMS remember) so, if the dimensions of the boot/trunk are expressed in inches by Ford and the dimensions of the cab are expressed in various different ways by the purveyers of the cabs, then of course there needs to be a standard scheme to express units of size, and then the conversions are easy. Don't forget, that's 100w RMS...
Probably the only XML seminar in the world where you get to go punting as well... Great thing is that it's been going for several years, and noone has fallen in the river yet, so the sense of anticipation is fantastic;o)
.. and the Dragon 32/64 crowd. Very similar architecture to the CoCo/TRS-80, some common peripherals (analogue sticks etc) and an identical version of BASIC, although the tape cassettes sadly weren't cross-compatible.
That'll be the reason. I wish I could get the hang of this intarweb thing.;o)
it's just very slow for me. slashdot loads very quickly though !
It is also possible that the subpoenas are invalid the FA is not clear on why they might be invalid - any insights ?
ok, I take it back, not slashdotted, just v.slow.
"They said it could be $750 per song. The letter said, though, that they could just pay $3,000, which would not be based on the number of songs." $3,000 == unlimited downloads.. hmm, not a bad deal... might be cheaper than 79 cents a track...
NCState webserver doesn't stand up to slashdotting
on
NC State Stands Up to RIAA
·
· Score: -1, Offtopic
so I won't get modded insightful for my frost piss (?) 'cos I couldn't rtfa
Slashdot Mirror
Actually many do. I'm in the media industry. Not only does my employer allow and encourage viewing listening of all manner of compressed and uncompressed content (let's not even talk about uncompressed content on 'Fister'), for many of my (000s of..) colleagues, it is a central part of their job. Next !
yes. the same servers that are able to stream 1500 HDTV channels to a single user, using multicast.
I'll drink to that ! I have JBoss tickling MADs fancy on a regular basis. Schweeet ;o)
funnily enough, no. so why exactly does a high latency low packet loss network slow down TCP based protocols so much ? Or are you saying that this doesn't happen ? I'm genuinely interested in your answer.
.. unless the last mile is vertically down from the satellite (at both ends of the link) ;o)
hmm, yeah, what I said - http://netlab.caltech.edu/pub/papers/FAST-ToN-fina l-060209.pdf
.. although I keep coming back to the sentence "...senses congestion by continuously measuring the round-trip time for the TCP acknowledgment and then monitoring how that measurement changes from moment to moment.".
I would imagine in the typical high-latency scenario, where regular TCP is mis-interpreting long RTT as link congestion, and backing off the rate, FastTCP is able to actually keep pushing the rate up, meanwhile keeping an eye on the RTT. I mean, the RTT shouldn't increase in line with the rate, unless the link actually *is* congested. So just increase the rate until the RTT increases, at which point you are genuinely maxxing out the link. I think that must be how it is working..
Yes, you read that right - 4 Libraries of Congress per hour !!!!
See http://www.fastsoft.com/research.html
Not so. High (>1500ms) latency *severely* affects TCP protocols like FTP. I encounter this on trans-continental WANs which go over satellite every day. I've tried some UDP compression such as FileCatalyst, and 15-20x speedup is possible on some links.
The problem is that "regular" TCP mis-interprets long Round-Trip-Time (aka latency) as link congestion and backs off the rate at which it is sending packets.
The bandwidth between point A and B may be rated at a high throughput, but TCP protocols such as FTP will never achieve that speed if the RTT is long. Increasing the bandwidth won't help !! So a slowdown of 20-30x is not uncommon on WAN links with high latency e.g. transcontinental, via satellite.
I've looked at technologies like Digital Fountain (and it's Java implementation, FileCatalyst) which use UDP and some clever mathematics to overcome latency, however it's not clear from TFA what FastTCP is doing underneath..
don't connect the dam floodgate controller to the internet ?
burble
:o)
+5 insightful, c'mon
Hey kickdown, while I like *totally* support what you're saying about shell, the AC does link to a 'send to power toy' where you need to press Ctrl to change the directory in the dosbox from the gui. Rock on d00d.
c'mon. logic says this is so unprobable that it will probably happen ;o)
Last year I attended an excellent seminar track on content and knowledge at this : http://www.xmlsummerschool.com/ - and one of the speakers had a great example - he wanted to be able to search for a guitar amp speaker cabinet that would handle the 100w (that's RMS) output of his Marshall amp, and fit in the boot/trunk of his car - I forget the make/model, let's say it's a Ford whatever... anyhow, the point is that the semantic search app would need to discover the dimensions of the guy's car's boot/trunk from Ford, then it could search for cabs that would fit in the boot/trunk that could handle 100w (that's RMS remember) so, if the dimensions of the boot/trunk are expressed in inches by Ford and the dimensions of the cab are expressed in various different ways by the purveyers of the cabs, then of course there needs to be a standard scheme to express units of size, and then the conversions are easy. Don't forget, that's 100w RMS...
Probably the only XML seminar in the world where you get to go punting as well... Great thing is that it's been going for several years, and noone has fallen in the river yet, so the sense of anticipation is fantastic
.. and the Dragon 32/64 crowd. Very similar architecture to the CoCo/TRS-80, some common peripherals (analogue sticks etc) and an identical version of BASIC, although the tape cassettes sadly weren't cross-compatible.
oh, for mod points. god bless you.
sadly, grub, I don't think that covers it..
http://news.bbc.co.uk/1/hi/world/americas/6560685. stm
that's one fucked up motherfucker. I can't begin to imagine how anyone related to this must feel.
.. don't forget the remote, a snip at 10 grand ..
it's just very slow for me. slashdot loads very quickly though ! It is also possible that the subpoenas are invalid the FA is not clear on why they might be invalid - any insights ?
so I won't get modded insightful for my frost piss (?) 'cos I couldn't rtfa