Forgot your password?
typodupeerror
Social Networks The Internet IT

Twitter Leads Social Networks In Downtime 175

Posted by timothy
from the great-to-be-top-of-the-list dept.
illectro writes "A study on site availability by monitoring service Pingdom shows that in 2008 Twitter greeted users with the 'Fail Whale' for more than 84 hours, almost twice as much as any other site. At the other end of the scale imeem and Xanga managed less than 4 hours of downtime for 99.95% uptime. Myspace, Facebook and Classmates.com were the only other sites studied which managed to stay up more than 99.9% of the time."
This discussion has been archived. No new comments can be posted.

Twitter Leads Social Networks In Downtime

Comments Filter:
  • 84 hours???? (Score:5, Insightful)

    by sunking2 (521698) on Wednesday February 18, 2009 @03:28PM (#26905615)
    And yet it had 0% impact on my life. So who really cares.
    • by gnick (1211984)

      Me either - I've never been tempted to visit twitter and would not have noticed.

      However, I have to give them props for the fail whale [designlessbetter.com]. I ran into that graphic somewhere-or-other and it's got to be the single best network-overload graphic I've seen.

    • by motek (179836)

      Well, 'social media' IS downtime. So either way, the users get what they are coming for.

    • Ten years ago, slashdot was the place where people would come and enthusiastically discuss hot new technology and trends.

      Today, the easy upmods come from playing the part of the crotchety old traditionalist who could not care less about whatever new thing those damn kids are into, because if you can't do it with an awk script and a soldering iron, it shouldn't be done!

      Oh slashdot, has your spirit died?

  • 84 hours?!?! (Score:5, Interesting)

    by RoFLKOPTr (1294290) on Wednesday February 18, 2009 @03:33PM (#26905659)

    I find it kind of strange that a site as incredibly simple as Twitter had so much downtime. Granted, they probably don't have the multiple dedicated redundant datacenters to their name like MySpace and Facebook do... but still, they're only serving little tidbits of text.

    • by LordKaT (619540)

      holy shit, it's RoFLKOPTr

    • plus a little bad luck and architecture trouble would do it. Unless of course you're thinking of Twitter as being a single apache/mysql pair.
  • by Thornburg (264444) on Wednesday February 18, 2009 @03:34PM (#26905681)

    If Twitter was the worst, with 84 hours downtime, one year is 8765.81277 hours, which means that Twitter was down .958268243% of the time. Not .9 (90%), but .009 (nine tenths of one percent). IOW, it has an uptime of 99.05%. Sure, that's not great compared to 99.95%, but it was down less than 1 in every 100 times you tried to reach it. I'm pretty sure Yahoo! doesn't manage that, and I know Microsoft's download servers don't manage that...

    • by Colonel Korn (1258968) on Wednesday February 18, 2009 @04:22PM (#26906467)

      If Twitter was the worst, with 84 hours downtime, one year is 8765.81277 hours, which means that Twitter was down .958268243% of the time. Not .9 (90%), but .009 (nine tenths of one percent). IOW, it has an uptime of 99.05%. Sure, that's not great compared to 99.95%, but it was down less than 1 in every 100 times you tried to reach it. I'm pretty sure Yahoo! doesn't manage that, and I know Microsoft's download servers don't manage that...

      Good numerical point, but Yahoo hasn't failed to load for me any time in the last 10 years, with something like 10-50 page views per day. Their uptime is thus no worse than based 0.99997 on my experience, which is means 300x less downtime than twitter.

    • If you drop to 98% availability...Jesus. It sounds good in a non-internet context, but the standard is 3 9's (99.9% uptime) at least. We're not talking 4 9's (99.99%, what you'd expect from your bank) here. We're talking about a site that's pushing 1 9. ONE. 98%!!!

      If they were your webhost, they'd be fired. It's just not an acceptable performance number for a big modern site.

    • Re: (Score:3, Insightful)

      by ari_j (90255)
      Frankly, I don't know how they did their measurements but Myspace gives me "an unexpected error occurred" often enough (and I only sign in when I get an e-mail notification of a new message or the like, to begin with) that it very much is expected.
    • Unless the site is more likely to go down during busy times, which would mean that the 1% of downtime it does have would be when a greater percentage (maybe 10%?) are trying to reach it. That would turn 99% uptime into an effective 90% - 95% uptime for their users, which is considerably worse. Having 99% uptime is great, but if the downtime comes when you're the #1 link on digg or slashdot, that means you lose a hell of a lot more traffic than you would have otherwise.
    • by tnk1 (899206)

      Depends on what time of day it was. If you had your 84 hours at some time where it was only daylight over the middle of the Pacific Ocean, you're probably okay.

      If you had all 84 of your hours during peak times, and event-driven times (major sports and news events), you just lost a lot of data. More importantly, you pissed off more people for being down 5 minutes than you did by being down for an hour at 4AM.

      Given that Twitter has issues that are well known due to capacity, my guess is that their downtime

    • Sure, that's not great compared to 99.95%, but it was down less than 1 in every 100 times you tried to reach it.

      Ahh, the old "99% is good enough" argument. Occasionally 99% is actually good enough but you have to be VERY careful with that argument. It is extremely easy to come up with examples where 99% is absolutely miserable performance. All you need is a large number of transactions or severe consequences for a failure. The former definitely applies here and possibly the latter too once money gets involved.

      For example if 99% were good enough reliability for air traffic controllers, each year 640,000 [natca.org] flights wo

      • Re: (Score:3, Insightful)

        The GP was saying that, all conversions be done, 84 hours is not as devastating as it can sound. That's not saying it couldn't, nor that it shouldn't, be improved.

        Mind you, it's a freaking social networking site. How many lives will be seriously inconvenienced (much less endangered) by its downtime?

        • 84 hours is not as devastating as it can sound.

          If Google was shut down for 3.5 days in 2008 it would have cost them $209 million [google.com] in revenue. A few more days of that and it might become real money.

          How many lives will be seriously inconvenienced (much less endangered) by its downtime?

          Lives won't but investments will be endangered and that does "seriously inconvenience" people. I present our current economy as Exhibit A. You think Google or Amazon or Yahoo would be the economic powerhouses they are if they were down for half a week per year? Customers wouldn't trust the service, investors would be wary, slashdot would bash them and they

  • by gEvil (beta) (945888) on Wednesday February 18, 2009 @03:35PM (#26905701)
    And here I am worrying about whether I should see my doctor after 4 hours of uptime.
  • I clicked the story expecting to see users waste more hours on twitter than other social networks. That would've been more interesting. The story above is... just plain boring.

  • by itsme1234 (199680) on Wednesday February 18, 2009 @03:47PM (#26905937)

    I think many of us recognize the potential power of twitter-like thingies. With this in mind I recently joined. It is beyond disappointing.

    - the site itself is barren, with basically no features - it is just like a '98 site in a bad way (not in a "Google-like" minimalist way)
    - can't get updates by SMS in Europe. OK, fair game, it isn't free. But you should be able to at least post by SMS, right? Somehow although they do offer local numbers (very nice) I wasn't able to actually verify any phone so can't update by SMS
    - they had updates by Instant Messenger as official feature for a while but couldn't make it work (why?! at least it should be practically free for them unlike SMS)
    - there are some 3rd party solutions to update by IM but none work (plus you have to trust the 3rd party)
    - same as above for updates by email

    So, yes, nice idea but poor execution.

    • Re: (Score:2, Informative)

      by dandv (1246510)

      - there are some 3rd party solutions to update by IM but none work (plus you have to trust the 3rd party)

      The Pidgin Twitter plugin [google.com] works [twitter.com].

    • I update it via identi.ca which I update via im. twitter is where most people are at that I communicate with so that is where my posts end up - but I don't ever go to the twitter site. It doesn't need to be more than it is.

  • by dandv (1246510) on Wednesday February 18, 2009 @03:55PM (#26906071) Homepage
    In an interview with RadicalBehavior.com, Twitter lead developer Alex Payne commented [radicalbehavior.com]:

    By various metrics Twitter is the biggest Rails site on the net right now. Running on Rails has forced us to deal with scaling issues - issues that any growing site eventually contends with - far sooner than I think we would on another framework. [...] At this point in time there's no facility in Rails to talk to more than one database at a time. [...] All the convenience methods and syntactical sugar that makes Rails such a pleasure for coders ends up being absolutely punishing, performance-wise. Once you hit a certain threshold of traffic, either you need to strip out all the costly neat stuff that Rails does for you (RJS, ActiveRecord, ActiveSupport, etc.) or move the slow parts of your application out of Rails, or both. It's also worth mentioning that there shouldn't be doubt in anybody's mind at this point that Ruby itself is slow. [...] I think it's worth being frank that this isn't one of those relativistic language issues. Ruby is slow.

    • Re: (Score:2, Interesting)

      by Anonymous Coward

      I mean wtf? This has been dubunked so many times.
      After this announcement someone wrote a plugin for rails that handled multiple databases.
      And you know, we had this huge ruby on rails application that never really took off. I would had really loved to have those performance issues they were describing.

    • by SanityInAnarchy (655584) <ninja@slaphack.com> on Wednesday February 18, 2009 @04:27PM (#26906541) Journal

      Running on Rails has forced us to deal with scaling issues - issues that any growing site eventually contends with - far sooner than I think we would on another framework.

      That is probably true. However, I would count that as an advantage -- better to deal with them sooner than later.

      At this point in time there's no facility in Rails to talk to more than one database at a time.

      There are many, many ways to talk to more than one database in Rails. In fact, it is possible to swap out the entire database layer of Rails and use another ORM, or no ORM at all. On the bleeding edge -- and Twitter might actually be a good candidate for this -- people have wired up Rails to CouchDB, which provides trivially scalable multimaster replication, and which, being HTTP, can be thrown behind any old load balancer -- which brings this back to a "just throw hardware at it" problem.

      All the convenience methods and syntactical sugar that makes Rails such a pleasure for coders ends up being absolutely punishing, performance-wise.

      Some of them do -- a good example would be Symbol.to_proc.

      However, Merb proves that this is not actually a Ruby problem, it is a Rails problem. And Rails and Merb are merging some point in the near future.

      It's also worth mentioning that there shouldn't be doubt in anybody's mind at this point that Ruby itself is slow. [...] I think it's worth being frank that this isn't one of those relativistic language issues. Ruby is slow.

      Somewhat true -- after all, Ruby 1.9.1 did double the performance of the language.

      But, relative to what?

      Turns out that, at least compared to other languages and frameworks (like PHP), Ruby is not slow [slideshare.net].

      It's also worth mentioning that while all of the Twitter alternatives may have enjoyed better uptime, they haven't had nearly the amount of traffic that Twitter does. We don't really know if they can scale -- but even supposing they can, Twitter was there first. And while they complain about those nice features being slow, they probably owe their success to those features for getting their product out the door faster than their competitors.

      It's also worth mentioning that this interview is almost two years old. Rails changes a lot in two years. In fact, Twitter were early adopters -- two years before that interview, Rails had only just shared commit rights. Two years before that, it didn't exist at all.

      It might be worth asking what version of Rails Twitter is using, and if they've noticed a change since then.

      • by tcopeland (32225)

        > better to deal with them sooner than later.

        Well said sir. Scaling is not about language opcode execution speed. Apps don't scale by eliminating a few JMPs and BEQs here and there. Instead, they scale by coming up with good _architectures_. Caching, partitioning, sharding, queuing, backgrounding... all that stuff. Given a good enough architect you could probably write EBay in Bash.

        • Not that those things are irrelevant, just that initially, they are less important.

          Certainly, it's possible eBay could save a lot of money on hardware by rewriting the site in a faster language (or even in C) if they started in Bash. It's also possible they would start looking at other optimizations -- all those little hacks you avoid during development could suddenly mean thousands of dollars saved.

          But you won't know which hacks are worth thousands of dollars, and which ones will cause thousands of dollars

    • does this mean metalhed77 wants to punch Alex Payne in the face? [slashdot.org]

  • by Sponge Bath (413667) on Wednesday February 18, 2009 @04:10PM (#26906305)

    I'm now posting on /. about Twitter.
    I live such a full life.

  • This is not all that meaningful unless you also completely correlate the uptime info with the number of users/requests/whatnot the site does.

    The report doesn't explore that sufficiently enough for me. I can make an app that has 100% uptime if it has one request an hour. Downtime is largely caused (directly or indirectly) by load, so in most cases downtime usually increases as user load (defined as user interaction and amounts of user data, and the actions of those users on that data) increases.

    Painting wi

  • Maybe there is, somewhere in the world, guy who pays lots of $$$ to ISPs to 404 Twitter site as often as possible. If this is the case, that guy is really rich and his brain works in similar way as mine.

  • by 222 (551054) <stormseeker@RABB ... minus herbivore> on Wednesday February 18, 2009 @04:25PM (#26906513) Homepage
    I no longer use Myspace (Thank god!) but it seemed like every time I tried to do something, I was redirected to an error page assuring me that their support staff would be notified...

    Sure, Myspace was able to display html in my browser, but it seems a bit far fetched to consider that "uptime".
  • Twitter is on par with Brangelina type stories. Find something more newsworthy to fill dead space with(!!) This is so frustrating.

  • by noidentity (188756) on Wednesday February 18, 2009 @04:55PM (#26907005)
    The social non-networking site I use, isolatr.com [isolatr.com], is never down, and has never failed to bring me zero annoying "friends". I highly recommend it.
  • by teknognome (910243) on Wednesday February 18, 2009 @05:01PM (#26907109)
    While I don't use twitter, it's downtime is bad enough (or people are obsessed enough) that not only is there IsTwitterDown.com [istwitterdown.com] but also IsIsTwitterDownDown.com [isistwitterdowndown.com]
  • these were what some people were saying in slashdot back a while ago about twitter, ruby and whatnot.

  • Don't like Twitter's downtime? roll your own [laconi.ca] and do better.

    (But honestly, I still don't see what all the hype about Twitter is. It's just a mashup between instant messaging and RSS from what I can tell, not sure why there needs to be a "service" wrapped around it.)

    • by Hadlock (143607)

      Twitter is RSS with SMS encapsulation. I'm pretty sure there's a way to express that entire concept in 3 letter abbreviations but I'm not going to wast the effort on it.

  • Yay fail whale (Score:3, Interesting)

    by bluefoxlucid (723572) on Wednesday February 18, 2009 @05:19PM (#26907435) Journal
    One of the most amusing error messages ever 3 Where the hell did it come from? Why is it flying with birds? What the fuck is this shit? Who knows! It's fail whale!
    • Re: (Score:3, Interesting)

      by revscat (35618)

      Birds tweet. They all tweet to each other. And they do so using tin cans and string.

      So they're flying along, happy tweeting on their retro iPhones when all of a sudden this doped up whale jumps in the middle of them, dragging them all to their doom.

      Birds = Twitter
      Whale = system/network load/myth of Rails scalability

      It's a gorgeous bit of iconography.

      • Myth of rails scalability? Dude, rails allows you to write most of your code without knowing what the fuck the code actually does. Maintainability lol.
  • We know Twitters architectural history, but anyone have a summary of the three big sites with the higher uptime? (Server-side of course). Commonalities would mean a lot I'd think.

    Sorry, I'm old and lazy or else I'd look it up myself.

  • ...are people who died from Twitter Recursive Downtime Syndrome (TRDS). More or less when Twitter goes down, they want to tweet about Twitter being down, when they realized how that makes them feel they want to tweet that, after about the third or fourth round of that, well it isn't pretty [asplode.net].
  • Join multiple sites. So when Twitter goes down, you can gripe on Facebook about it, and when FB goes down, you can tweet away.

  • So myspace hardly ever goes down. They still to this day have issues of not having enough resources for the amount of traffic they get (may be part of the reason that most people I know have all but ditched myspace and went to facebook). Look, if there Are so many people on your site every single night that some people get DOS, and those who do get in cannot do anything, while they have the brilliance to take down resources during peak hours to install untested site updates, then what difference does it mak

Life. Don't talk to me about life. - Marvin the Paranoid Anroid

Working...