Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
The Internet

Net Traffic Shocks Mimic Earthquakes 157

Posted by timothy from the aftershocks-and-all dept.
belphegor writes "Technology Research News is running an article describing research linking the similarties between Earthquakes and Internet traffic. By pinging hosts across the network, researchers 'were able to measure frequent changes in Internet congestion...results showed that the Internet, like the earth's network of faults, exhibits criticality -- a condition of sudden and drastic change. "Sudden drastic congestion leads to a large value of the round trip time of the ping signal, which is identified with a main shock," said Abe. The researchers referred to these sudden, drastic traffic changes as Internetquakes.' They also saw 'aftershocks' that can be mathematically described in a similar manner to the seismic ones more familiar to many Californians."
This discussion has been archived. No new comments can be posted.

Net Traffic Shocks Mimic Earthquakes More

Net Traffic Shocks Mimic Earthquakes

Comments Filter:

  • cool (Score:2, Funny)

    by Anonymous Coward
    maybe we'll get lucky and if we ping the .NET side enough it will fall into the ocean! :)
  • Insert tastelass Slashdot-effect joke here.

  • The name for this phenomenon? (Score:4, Funny)

    by viper21 ( 16860 ) <scottNO@SPAMiqfoundry.com> on Tuesday August 27, 2002 @01:35PM (#4150552) Homepage
    The Slashdot Effect.

    I'm surprised that it isn't mentioned in the article. They are probably trying to patent it, I'm sure.

    -S
    • I think Slashdot should rent out its services to companies that want to test their server's stability. They can make a new topic like:

      "unicron writes in to tell us he's found a site with over 100,000 free, hi-rez pr0n pics, and it's all completely free! You can even download them all as one big-assed compressed file!"
      • "unicron writes in to tell us he's found a site with over 100,000 free, hi-rez pr0n pics, and it's all completely free! You can even download them all as one big-assed compressed file!"

        This would be known as the Slashdot effect.

        "slagdogg writes in to tell us he's found a site with over 100,000 free, hi-rez pr0n pics of Natalie Portman, and it's all completely free! You can even download them all as one big-assed compressed file!"

        THIS would be an Internetquake!
  • So what does a Slashdotting rank on the Richter Scale?
  • Slashdot Effect = San Andreas Fault?

    It's interesting, I'm glad to see the article ties it in with other complex systems as well.

  • Seems like /. has been responsible for many a 'netquake. Now /. will crash your web servers FOR SCIENCE!

    • Slashdot has also been a mitigating factor to some of the most severe 'quakes', especially if you're discussing the information transmitted rather than just network traffic.

      Case in point? 9-11-01.

      CNN was gone. USA Today was gone. Fox News was gone. Slashdot, however, was providing first-hand accounts of the disaster/attack as well as discussion about them.
  • NO front page postings during the world series this year!
  • Causing plate shift all over the internet :)
    sorry, it was just too easy to let go by :P
  • apparently Technology Research News will be the next website to feal this 'internet earthquake'...

    maybe with followup stories it can have some 'aftershocks'...
  • I've lived in So Cali for two years, and have not felt one earthquake. I didn't move here specifically to be in earthquakes, but still I feel cheated :(

  • That explains a lot (Score:4, Funny)

    by r_j_prahad ( 309298 ) <r_j_prahad@hotm a i l .com> on Tuesday August 27, 2002 @01:39PM (#4150605)
    Our company's webmaster has managed to exactly duplicate the San Andreas fault then. That box is shakier than California... oops, here comes another tremor now. Yup, reduced to rubble one more time.

    One of these days I'll find some way to get the URL posted to Slashdot and I can see if computers really do burst into flame and shoot out showers of sparks like on the old Star Trek....
  • I wonder what a good Slashdotting would look like? Probably 11 on the Richter scale.

  • This behavior is predicted by Celullar Automata (Score:4, Interesting)

    by eyefish ( 324893 ) on Tuesday August 27, 2002 @01:41PM (#4150614)
    This pattern of behavior (where two completelly different things show the same underlying behavior and/or explanation) is exactly one of the things that Stephen Wolfran is trying to explain in his book A New Kind of Science [wolframscience.com] (see amazon link and reviews here [amazon.com])

    Basically, everything in the universe can be explained as a huge network of nodes, where all these node do is computations following very simple rules. From such simple rules we get all the laws of physics, human behavior, chaotic behavior, and in this case the behavior of an earthquake and Internet traffic.

    • Re:This behavior is predicted by Celullar Automata (Score:4, Informative)

      by WEFUNK ( 471506 ) on Tuesday August 27, 2002 @02:08PM (#4150844) Homepage
      Another book that is more directly related to this behavior is Per Bak's "How Nature Works". Bak demonstrates how many systems in nature, including earthquakes, highway traffic, and evolution, are governed by very similar power laws and he proposes that these systems naturally evolve to a state of "self organized criticality" as is mentioned in the article.

      A simplistic model that he uses to describe this principle is a conical pile of sand that is built up by dropping one particle at a time. The pile will build and build and then experience surface avalanches apparently set off by only a single piece of additional sand. The frequency and size of these also follow the same type of power law. Self organized criticality is really a updated version of catastrophe theory and theories involving constraints. I'm still not sure if Wolfram has really shed any new light on this area, but some of his work might be used to explore the fundamental cause behind this apparently common principle.

      I believe that some of these ideas are already used quite practically. For instance, knowledge that traffic jams can be caused without any external cause (accidents etc.) can provide insight into whether changing speed limits or adding additional lanes can ease congestion (or worsen it). Models and simulations can be built and sometimes the answers are found to be contrary to common sense.
    • This should be -1 Pseudoscience. There is no evidence to back up any of the theories in that book. It's one of those "look at results, form theory to fit" books.
    • Right here [wired.com]

      Although it's pretty vague like many Wired non-technical articles.

      But i find it amusing to see a person spend 10 years on such idea while running a company.
      Also talks about his life, pretty interesting :)

    • You should be warned that using the strings "Stephen Wolfram" and "A New Kind Of Science" in the same sentence might make you look like a clueless idiot.

      These guys [memepool.com] can tell you more about the anti-Wolfram position.

  • the net is a non-linear resonant circuit (Score:5, Interesting)

    by cats-paw ( 34890 ) on Tuesday August 27, 2002 @01:41PM (#4150616) Homepage
    I like to think of the net as a non-linear resonant
    circuit. The correct starting conditions give
    you very large responses to "small" inputs.

    It's resonant so you see "ringing" in the response.

    Seems to me that you can definitely look at the
    earth's crust in a similar manner.
    • Hmmm... I like the idea of an Internet that "rings". But isn't it the "ringing" that we would like to get rid of from the net, in the same way we might wish the earth didn't "ring" so well during earthquakes?

      What is the opposite of this model? Something like a marshmallow or ball of silly putty, that deforms to store energy and allows that energy to be released over time? I know that might affect the individual round-trip times, but it might improve efficiency on a larger scale.

      I'm not sure what the rules for the nodes would be, but obviously there would have to be a way to balance "slack" and "stress" at different scales simultaneously, in order to avoid criticality. Anyone know about any models that have this kind of behaviour?
  • Shut up,
    already! Damn!
    Don't wait for your neighbor,

    Green Eggs and Ham!

    Doin' the netquake!!

    You know what I'm talkin' about?

  • post a stupid slashdot effect joke, looks like about 20 people beat me to it, dammit! Oh well here is a different stupid joke:

    Q: What has 4 legs and smells like fish?
    A: Bill Clinton's desk

    • Offtopic? This was just as funny as the 18,000 /. effect jokes. Here's a clue for all future posters - if it is obvious to you, then it is obvious to everybody. You are *not* thinking of something that just happened to slip by 300,000 other nerds. Don't post it.

      Seriously, give this guy a break for trying to be original.

  • ... I don't think earthquakes are an entirely appropriate metaphor. I'd think of it as more like dropping flaming bombs. Then, like in Populus, a flaming dude goes running back home only to set it on fire. Then people run out of it into neighboring homes and set them on fire. And then..

    Well shoot now I wanna go play Populus. I don't feel like fleshing out my point.
  • People get money from the government to research this kind of stuff? I want a government grant. I want to compare the internet to the growth of fungus on oranges.
    • People get money from the government to research this kind of stuff? I want a government grant. I want to compare the internet to the growth of fungus on oranges.

      My theory with both the orange fungus and the internet is this:

      The more disgusting it is, the more you can charge people to watch it.... :)

    • Better yet, smear the fungus on a canvas, cover it with feces and religious symbols, and get a lifetime fellowship (read: free money) from the U.S. National Endowment for the Arts.
  • I've always referred to earthquakes as a "geological slashdotting". I think that this backs up my theory.

  • Gamers have known about the Quake(s) and (low) ping time relationship for years...

  • This would, of course, make /. the san andreas fault of the internet world. Causing havoc and destruction to small, unprepared villages up along the faultline.
    • /.'s more like the entire "Ring Of Fire" that the San Andreas is part of, even international sites get flattened by the seismic shocks that are the voluntary DDoS by thousands of geeks everywhere.

      Maybe that should be "Token Ring Of Fire..."
  • I'm not surprised about this behavior: after all, the Net has been mostly created in California, so there must have been some subconscious influences on its design so that it mimics the quakes that plague the area. :)
  • I see a movie coming out of this.
  • Cool!

    Now all of the money the government has spent on the detection of earthquales can actually help ME. Advance knowledge of net connection would make my life much easier.

    Heck, VOIP might even actually be commercially feasable.

  • Is it just me or is this crap really stretching it. I think with 10 years and a good budget, I could write a convincing enough paper describing how cow feces is similar to chocolate.

  • Boy, this must have really made those guys leap out of bed in the morning, eager to get to work!
  • "Sudden drastic congestion leads to a large value of the round trip time of the ping signal, which is identified with a main shock,"

    So what you are saying, is that when there is traffic, things slow down??? and i thought that it was cause someone [insert funny analogy here]

  • Comment removed based on user account deletion
  • does that mean that instead of saying:
    "in 1996, kobe, japan had a terrible earthquake"
    we should say:
    "in 1996, kobe, japan got earth slashdotted" ?

    or "for san francisco in 1906, the fire after the earth slashdot effect was more damaging than the orignal earth slashdot itself."

    or "accompanying large earth slashdottings, there often follows many smaller earth afterslashdots."

    are we to replace the richter scale with cowboy neal poll results as well? ;-P

  • Fault lines (Score:2, Funny)

    by eclectus ( 209883 )
    and the obligitory Windows Joke(tm)....

    I guess that means all the WinNT servers represent the fault lines?
  • Quake is responsible for all the slow-down on the Internet?!? For some reason, i thought that Counter Strike [counter-strike.net] was more popular...
  • When ever a new pentium 4 CPU is released Germany experiences sudden lag.

    When ever new office toys are sent out for review Australia takes a hit

    and when ever some idiot with a backhoe digs up a backbone line, well hell, the entire USA goes ploink.
    • *COUGH* fixs HTML *COUGH*

      When ever a new pentium 4 CPU is released Germany [tomshardware.com] experiences sudden lag.

      When ever new office toys are sent out for review Australia takes a hit [dansdata.com]

      and when ever some idiot with a backhoe [cat.com] digs up a backbone line, well hell, the entire USA goes ploink.
      • Would it have killed you to find a story about a guy actually digging up a backbone? It's had to of happened somewhere.

        I'd appreciate it if you could find a few dozen of them, because I'm trying to convince myself that my Cable-Modem couldn't possibly be this spotty with it's service, so it's gotta be something like the backbone breaking somewhere.
  • Needless to say, 90% of all these Internet quakes are the result of a link on Slashdot. (And 95% of all statistics are made up.)
  • How much of that criticality is due to the Slashdot Effect?

    .
  • The researchers referred to these sudden, drastic traffic changes as Internetquakes.

    Why? We've been calling it the Slashdot Effect for years. :-)

  • Please make it stop... (Score:4, Insightful)

    by SkyLeach ( 188871 ) on Tuesday August 27, 2002 @01:52PM (#4150729) Homepage
    "linking the similarties between Earthquakes and Internet traffic..."

    I thought that there was nothing else they could call "like a <insert something to do with technology>", but there they go one-upping me again. So our brains are like computers, our genes are like source code, and our networks are like geology. Perhaps we really are in a matrix? Or perhaps all those similies in school are finally showing their ugly far-reaching effects on society.

    We should commission a study, but then, somebody probably already has and I'll read about it on slashdot next week.

  • Slashado (Score:2, Interesting)

    by Kristoffor ( 562485 )
    I see the /. effect more like a tornado than an earthquake. An impenetrable wall of traffic is created around the site which sit in the calm "eye" of total disconnection.
  • it's amazing how these scientists still have the balls to go and feed press with research results like this. You know, in certain situations you could find empirical proof to link for example your toilet visit to network congestion.
  • Seemingly irrelevant car fires occurringly safely off the side of the highway back up traffic for miles as traffic around the incident slows to a crawl. This behavior mimics that of seismic activity and internet traffic slowdowns.

    Internet experts now warn you in the event of a car fire, to stand in a doorway or underneath another heavy structure.

  • Seems... (Score:2, Interesting)

    by Hayzeus ( 596826 )
    ...like a lot of work to do to come up with what basically appears to be not much more than a metaphor.
  • DUH!

    Stuff breaks and because it's a network other stuff breaks too.

    wow.
  • Why back in the late 20th century, we had a series of netquakes we all blamed on SprintNet. It was a horrible time and the suffering was great, many a word was uttered in disgust, and a few out-loud. Why, the earth practically stood still as if we was all trapped in a limbo. When a traceroute succeeded, through Northern Michigan University, to Canada, to Rochester NY, then on through New York City and Maryland, then down to New Orleans, back up to St Louie and off to Kansas, we all knew there was a serious catastrophe around Chicago and the moanin' of lost souls was great. Eventually they recovered, but we loathed them old Cisco routers that SprintNet used, they was just itchin' for a major disaster. We'd pick up the peices, o' course, as what misfortune befell us didn't matter to they'uns. Them was tough old times. Your DoS attacks ain't never come close. 'Course we then got saddled with CIESIN and was sharing a T1 with another school, by gum, it could make a feller cry.

    You think you has it good, but all we needs is a few WorldComs to twitch the switch and ya'll be right back there in '93. Yep.

  • NEWSFLASH... (Score:5, Funny)

    by thrillbert ( 146343 ) on Tuesday August 27, 2002 @01:56PM (#4150763) Homepage
    Internet, Aug. 27, 2002 - The Technology Research News (www.trnmag.com), which recently wrote an article on how Net Traffic mimics earthquakes, experienced an earthquake that shattered their networks, and brought down their servers. The earthquake measure 8.4 on the CmdrTaco TEBCF scale (time elapsed before complete failure).

    News Correspondent Kimberly Patch had this to say via her cell phone connection: "one minute we're up and working, next thing you know everything just went crazy. People were screaming and running out of the building. I tried to keep calm, but at one point, knowing the magnitude of things, I paniced and ran out of the building screaming bloody murder!"

    A representative from OSDN was quoted as saying: "heh.. heh.. umm.. heh heh.. nice huh?".

    Calls to slashdot.org were not immediately returned.

    ---
    (Score: +1 Funny, +1 Interesting, +1 Too Much Time on Hands)
  • I wonder what percentage of the traffic in the networks they were "monitoring" was the actual pings? "Dang. Seems to happen every time we do this experiment!" Just curious...
  • After pinging (naughty me) www.trnmag.com I have found a strong resilience to strong seismic activity such as the slashdot effect. Though there seems to have been a quake near the end. And some gaps may have appeared since there was a 5% packet loss.

    PING www.trnmag.com (209.238.138.241): 56 data bytes
    64 bytes from 209.238.138.241: icmp_seq=0 ttl=231 time=268.2 ms
    64 bytes from 209.238.138.241: icmp_seq=1 ttl=231 time=266.4 ms
    64 bytes from 209.238.138.241: icmp_seq=2 ttl=231 time=267.6 ms
    64 bytes from 209.238.138.241: icmp_seq=3 ttl=231 time=266.0 ms
    64 bytes from 209.238.138.241: icmp_seq=4 ttl=231 time=267.2 ms
    64 bytes from 209.238.138.241: icmp_seq=5 ttl=231 time=262.1 ms
    64 bytes from 209.238.138.241: icmp_seq=6 ttl=231 time=263.4 ms
    64 bytes from 209.238.138.241: icmp_seq=7 ttl=231 time=264.7 ms
    64 bytes from 209.238.138.241: icmp_seq=8 ttl=231 time=265.9 ms
    64 bytes from 209.238.138.241: icmp_seq=9 ttl=231 time=266.7 ms
    64 bytes from 209.238.138.241: icmp_seq=10 ttl=231 time=268.0 ms
    64 bytes from 209.238.138.241: icmp_seq=11 ttl=231 time=267.4 ms
    64 bytes from 209.238.138.241: icmp_seq=12 ttl=231 time=265.7 ms
    64 bytes from 209.238.138.241: icmp_seq=13 ttl=231 time=265.7 ms
    64 bytes from 209.238.138.241: icmp_seq=14 ttl=231 time=267.5 ms
    64 bytes from 209.238.138.241: icmp_seq=15 ttl=231 time=265.0 ms
    64 bytes from 209.238.138.241: icmp_seq=16 ttl=231 time=263.3 ms
    64 bytes from 209.238.138.241: icmp_seq=17 ttl=231 time=268.2 ms
    64 bytes from 209.238.138.241: icmp_seq=18 ttl=231 time=265.7 ms
    64 bytes from 209.238.138.241: icmp_seq=19 ttl=231 time=267.0 ms
    64 bytes from 209.238.138.241: icmp_seq=20 ttl=231 time=265.1 ms
    64 bytes from 209.238.138.241: icmp_seq=21 ttl=231 time=262.2 ms
    64 bytes from 209.238.138.241: icmp_seq=22 ttl=231 time=530.4 ms // small leading quake
    64 bytes from 209.238.138.241: icmp_seq=23 ttl=231 time=1168.9 ms // main quake
    64 bytes from 209.238.138.241: icmp_seq=24 ttl=231 time=266.6 ms
    64 bytes from 209.238.138.241: icmp_seq=25 ttl=231 time=267.8 ms
    64 bytes from 209.238.138.241: icmp_seq=26 ttl=231 time=266.7 ms
    64 bytes from 209.238.138.241: icmp_seq=27 ttl=231 time=264.4 ms
    64 bytes from 209.238.138.241: icmp_seq=28 ttl=231 time=265.8 ms
    64 bytes from 209.238.138.241: icmp_seq=29 ttl=231 time=709.7 ms // after shakings
    64 bytes from 209.238.138.241: icmp_seq=30 ttl=231 time=450.5 ms
    64 bytes from 209.238.138.241: icmp_seq=31 ttl=231 time=462.5 ms
    64 bytes from 209.238.138.241: icmp_seq=32 ttl=231 time=542.6 ms
    64 bytes from 209.238.138.241: icmp_seq=33 ttl=231 time=888.5 ms
    64 bytes from 209.238.138.241: icmp_seq=34 ttl=231 time=263.9 ms
    64 bytes from 209.238.138.241: icmp_seq=35 ttl=231 time=268.2 ms
    64 bytes from 209.238.138.241: icmp_seq=37 ttl=231 time=267.1 ms
    64 bytes from 209.238.138.241: icmp_seq=38 ttl=231 time=262.4 ms

    --- www.trnmag.com ping statistics ---
    40 packets transmitted, 38 packets received, 5% packet loss
    round-trip min/avg/max = 262.1/341.9/1168.9 ms
  • They're trying to take The SlashDot Effect(tm) out of our vernacular!

    On the flip side, once that happens we can start talking frankly about P2P.
  • Let's say that the Internet IS like Terra's fault network. Then this means the "Ring of Fire" would look something like this:

    Washington, USA: Duh. Though I know it's overstated, it has to be Microsoft. If you equate them to Mount St. Helens, though, it's not really that bad. Occasional hiccups, not too terrible.

    Japan: Either Mt. Fuji is the tons of crazy pr0n, both human and Anime, or the congestion that the Sony PS2 and Nintendo GameCube will cause when MMO gaming hits consoles full force.

    The Philipines: Don't even get me started. If I had a nickle for every time I say the words "pinoy" and "pinay" on IRC, I'd be rolling in cash like /. is in AC posts.

    Hawaii, USA: We consider it something of an expensive paradise, right? Compare that, and the archipilego's many volcanic spots, to the P2P MP3 revolution. While it gurgles, it never really erupts (RIAA and MPAA). The sons of Napster are living it up while they can.

    California, USA: One word, three times: Pr0n, pr0n, pr0n! Where is most of your congestion? Why, in the state that's doomed to fall off the continent and into the vast warm waters of the Pacific, of course! How much of it is made there? It's Hollywood, after all. That place gives a whole new meaning to the word "fault," eh?

    Booby Prize Winner - Russia: While not really a big part of the "Ring of Fire," Russia has it's own explosive history, that being the Tunguska Blast in the early 20th Century. Open Source anyone? It's not so much congestion (though this is where "/.ings" come in) as it is the explosion that OSS has enjoyed as of late.

    And they got government money to do this sort of crap? No study...
  • perhaps a large upswing in the volume of people sending out useless ICMP traffic to measure this "phenomenon"? ;^)
  • So now when we hit a roadblock on the Information Superhighway and have to sit through buffer to buffer traffic, it's due to an Internetquake?
  • " In the case of the Internet, each site is connected to the whole through servers, which host web pages, and routers, which coordinate traffic. The servers and routers themselves also form a hierarchical clustering structure, said Abe.

    So these server items Host only web pages, and all the traffic on the internet is web traffic, which moves from link to link by a "router". This is very upsetting that college educations and many many years of research show something that could be acomplished in 20 mintues with a SDSL line, a p0rn site on a connection faster then the SDSL line, and finally root@box'#ping xxx.xxx.xxx.xxx

    I wounder if they have their MCSE?

  • Predicting Congestion? (Score:3, Insightful)

    by Anonym1ty ( 534715 ) on Tuesday August 27, 2002 @02:19PM (#4150939) Homepage Journal

    Well what about the problem in this self organizing system, that you probably change the inherent nature of the system by observing it?

    "...The simple ping signals were emitted every second and traveled through 10 different routers before the signals eventually reached the destination computer. The researchers were able to measure frequent changes in Internet congestion by measuring the time it took a series of signals to complete a round trip..."

    How much bandwidth will these test pings take up inorder to have enough data to construct a model of current conditions that's good enough to predict bandwidth changes far enough ahead to make smart routers that can work around the congestion prior to it's existance? It would be kinda neat to have an internet that routes around inorder to prevent predicted congestion.

    • I don't know how accurately you can correlate simple congestion estimates with ping times alone. Obviously, the small size of ping packets vs. the larger mltipacket nature of other messages creates a difference. And the ping times alone show large variablity even on a closed network with just two machines.

      Do you remember the article about estimating the speed of light by measuring the ping return times over varying lengths of CAT-5 cable on a simple loopback line? I replicated that sort of thing on my own two machine disconnected-from-the-net network and also saw the same sorts of 1/f distribution of ping times when I ran ping with a time delay of 0.1 seconds. If you run ping with a 0.0 second delay, flooding the system, you see a sort of weibull distribution (a skewed gaussian). If you run ping with a default of 1 sec and 10k iterations, you see an almost gaussian distribution (more accurately, a very slightly skewed weibull).

      But if you look a the power spectra of these distributions, the delay factors in the ethernet switch or card or PCI bus or software or the kernel delays from the interrupts continuously occuring in the background all get together to create delays that could be mischaracterized as congestion, if only I didn't already know that this network consisted of two machines and that the only traffic on this network was composed of pings and responses.

      btw, I was getting averages of 1.4 - 1.7 msec, with stds of 0.026 msec using cables of 5 m, 10 m, and 300 m (meters) length.
  • I can't help but wonder about how many of these events were due to employees in a NOC somewhere downloading all the pr0n they could eat over an OC-48? :-)

  • Predicting Lag (Score:5, Interesting)

    by tarsi210 ( 70325 ) <nathan@nat h a n p r a l l e . com> on Tuesday August 27, 2002 @02:29PM (#4150994) Homepage Journal
    I always wondered how long it would be before someone tried to give a "weather report" for the Internet. I imagined that some day the daily Internet traffic report would appear right alongside Reporter Bob up in TrafficCopter 7 reporting on the condition of Highway 69 at 8am.

    "Well, Joan, it looks like Slashdot it at it again. You'll want to keep clear of Alter.net this morning as it appears to be having congestion problems around the Midwest. Communications to anything near a corn field is likely to be slow to smack-me-dead stupid through the 9 o'clock hour. Queue up those emails, folks."

    One has to wonder if it's even possible to predict and gauge incoming traffic problems. I guess you would have to know the effective capacity of the Internet, and sub-portions of it, at all times. I can see how a router's effective capacity could be measured by its effective throughput and cache. Your "sentinal level" would occur when the cache is full and bandwidth is maxed and the packets start to get a one-way trip to /dev/null. However, I would think it'd be tough to always keep up on what everyone has for capacity unless the backbone providers regularly publish these stats, as well as the stats on current traffic.

    I don't know if it would be helpful or not. One one hand, it'd be handy to know that the reason I'm getting 1900ms ping time to SF this morning is because some dumbass tripped on a power cord, but on the other hand, if I really care that much, I can probably figure it out using traceroute et al.

    Maybe it's a solution to keep everyone and their dog from flooding the 'net each time a router bites the big one and makes a suburb blink out of existence for a few hours. Other than that, it just sounds like a good excuse to draw pretty graphs.
    • I always wondered how long it would be before someone tried to give a "weather report" for the Internet

      You mean like this? [mids.org]
      • All well and good, but:
        • Most online "weather" reports kinda suck. One shows all routers in the world down right now, and the one you linked to is difficult at best to understand. I'd like a little better graph if you're going to give it to me visually.
        • I was talking more of a regular, syndicated, CNN News Today Internet weather report or something. Something that, because so many people are affected by the Internet, would regularly tune into and watch...like the weather channel or something.
  • They also saw 'aftershocks' that can be mathematically described in a similar manner to the seismic ones more familiar to many Californians."

    Or, more simply described as Slashbacks.
  • The "big one" will be caused by a site called -
    "Nude Brittney Spears advocates Linux, Star Trek, over Windows, Star Wars."

  • Faulty Analogy? (Score:2, Interesting)

    by obnoximoron ( 572734 )
    In both the internet and the network of earth faults, there are interactions between elements on multiple scales in time and space. That analogy is not hard to guess, but these physicists seems to have contributed a systematic measurement and modeling study. Fine.

    So far, just from the reading the press article, my doubts about their work are twofold:

    1. their ping frequency is one ping per second. This is a poor sampling of the time series of packet delays. Specifically poor sampling could lead to a false alarm or a miss. False alarm meaning that the sudden drastic 'congestion' could very well be an isolated event suffered by the ping packet or its adjoining ones; the rest of the packet delays in the one-second interval may be small. 'Miss' meaning you didnt catch the congestion due to the poor sampling. The ping packet got through quickly but there were congestion delays suffered by the rest of the packets in its one-second interval.
    In short, ping times are a poor indication of the true traffic status of an IP network. One ping per second is especially pathetic if one of the bottleneck links in your route is a high-speed OC-3 or higher link. For example, at 155 Mbps (OC-3), there can be upto 13,000 typical IP data packets through the router in one second. Only one of these is the ping packet!

    2. Their flawed analogy between fault energy and congestion level. Fault energy builds up slowly over years and is then released suddenly in an earthquake. They compare to this user sessions running over hours and then suddenly contributing to short periods of drastic congestion. So are they saying that congestion builds up slowly over hours at a router or server as more and more packets are processed? Ha! I am still ROFLing on this one!

    There research will vastly improve if they collaborate with internet researchers and engineers who have a working knowledge of the internet's routing and flow control protocols, instead of approaching it from a purely black-box modelling point of view.

  • So basically what they're saying is that the Earth is one giant computer? [douglasadams.com]

    ::Colz Grigor

  • "exhibits criticality"

    Is as much a mimickry of earthquakes as it is of farts, naps, and almost running into that hot chick in the hall.

    --Blair

  • ...Earthquakes and the Internet are apparently very different...

    These people should be in rocketry.

Slashdot Top Deals

How much net work could a network work, if a network could net work?

Close