Twitter Leads Social Networks In Downtime 175
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."
84 hours???? (Score:5, Insightful)
Does it make that much difference? (Score:5, Insightful)
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...
Re:84 hours?!?! (Score:1, Insightful)
Cause they picked ruby on rails, and built it in a totally unscalable way.
Also, who cares.
Re:'Get a life' as a positive suggestion (Score:2, Insightful)
Demands? Facebook requires you to update? You must be using a different site than the one I know.
Re:84 hours?!?! (Score:2, Insightful)
Awww, the poor little rails fan is gonna get violent on the internet. Good way to make a point, hotshot.
Re:'Get a life' as a positive suggestion (Score:4, Insightful)
No kidding. I've always been frustrated with people who claim they're "deleting my account because of the amount of time I waste on here". Hello? Just stop spending so much time on it...
The only reason I can think of to delete your account would be if you actually wanted to mass-delete every note and posted item you'd posted, every post on your wall, and every tag that you'd ever been given. Otherwise, just disappear for a week or three. Your "friends" will forgive you, and the real ones might even call or e-mail if they're really that concerned about you falling off the face of the earth.
Re:Does it make that much difference? (Score:5, Insightful)
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.
Re:84 hours?!?! (Score:3, Insightful)
Oh BS. A huge volume of extremely easy data. No images, no War & Peace length text posts. Just a lot of short, sweet, and simple text.
I want you to say with a straight face that it's really just the amazing volume of data that separates a highly reliable and available site like Facebook from a constantly failing jumped up IRC client like Twitter.
Twitter is a dog. And because it's written in Rails, it's a special needs dog that has to go to the vet a lot.
What is the uptime metric? (Score:4, Insightful)
Sure, Myspace was able to display html in my browser, but it seems a bit far fetched to consider that "uptime".
Re:84 hours?!?! (Score:3, Insightful)
Re:Does it make that much difference? (Score:3, Insightful)
This is nonsense. (Score:3, Insightful)
The problem isn't people obsessed with Rails... the problem here is people who just don't like it, for whatever reasons of their own. Well, your reasons *ARE* your own. Please keep them to yourselves unless you can start coming up with facts rather than unfounded insults.
Quote from Twitter representative: "I strongly believe that the best tool for the job is the best tool for the job. Rails is the best web application framework around for rapid prototyping and, as aforementioned, building CRUD-style applications. I would choose Rails again for such a project."
Twitter stated that they simply did not plan ahead for the popularity of their service. Period. That is not the fault of the platform they use.
Re:This is nonsense. (Score:2, Insightful)
Well, since we agree that the Twitter people are incompetent, why should their opinion matter? Maybe their love of Rails is the root of the whole problem, and they're just to wedded to the environment to see it.
I personally think it is part of the problem. Development/deployment frameworks add a non-trivial amount of overhead, which is something that cannot be spared on a high volume applications.
Aside from all that, I just love tormenting Ruby fanatics. They're as defensive and strident as any C geek, though, unlike the C geeks, Ruby/Rails people can't point to any performance increases to justify their fanaticism.
Re:84 hours?!?! (Score:2, Insightful)
I'd point out that your defense of rails is just as silly as the parent's attack. "Tons of successful rails deployments" adds exactly zero datapoints to the argument.
And I'm not going to accept the statement of the Twitter folks that they would use rails again as some kind of argument in favor: a) they've failed, so their decision-making record isn't stellar and b) obviously they're going to defend their decision, since saying "Oh, we screwed up" is practically unheard of these days.
This isn't a troll (really!) I'm just pointing out that I think you and the parent are not really having a productive discussion of the merits/demerits of rails. (Though what do I expect, this is /. after all) :-)
Re:84 hours?!?! (Score:3, Insightful)
Well you misunderstand scalability. Scalable DOES mean you can just throw more hardware at something, as you said. But you omit the part the scalability has limits. There's always a bottleneck, so no app is infinitely scalable. You take your best guess at what its intended audience will be and you proceed form there.
You can say my app's been designed so that it will scale when n is between 1 and 100,000 for instance. Or that it'll scale between 1 and 100,000,000,000,000. But the app you design for the later will be very different internally and take a lot more time to design. Most apps start out working only for small n because most apps will never get to big n or if they do, they do so gradually and give developers time to deal with it. Twitter was cursed by its own success, they didn't know it'd be as popular as it turned out.
Let's look at some specific technical details.
For most sites a single master DB server with slaves for reads is fine. Most apps out there don't need multi-master capabilities and thus implementing that would be a waste of time. In fact, most web apps you see out there are designed to run on a single DB server for reads and writes.
When Twitter started that's what they had and it worked. Then twitter got unexpectedly popular, and at that point in time Rails did not support multiple DB connections per instance, so sharding wasn't really possible. Days after they announced this problem it was fixed so it's no longer an issue.
We could speculate on problems they've had since then, but that would just be speculation. They've said repeatedly however that their stack's scalability issues have been in the non-rails parts (like the DB). Since they're the only ones who actually know what's going on, I'm inclined to believe them.
Let's look at some other social app failures,
An example would be myspace, which had horrible problems scaling, even though that was written in PHP which you consider to be super scalable. Or we could look at friendster, which was written in enterprisey java, which also had huge scaling problems. It's not the tool, it's how you use it.
Additionally, when you scale real world systems you just may find out the bottlenecks aren't where you thought they'd be during the design and testing. Runaway successes like twitter are vulnerable to this fact.
If you want to fault the Twitter engineers you really can only fault them for bad judgement in estimating the size of their user base. I don't think you know what Rails actually is. It's just helpers for YOUR code. Ultimately, a rails site is mostly your code, and your architecture.
Re:They are cut off (Score:2, Insightful)
What the readers of the feed think: Having a bunch of feeds to follow is a mildly amusing way to kill time
What the owner of the feed thinks: That he's so awesome/important that people want to know what he's doing at all hours
Re:99% isn't good enough (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?