Archive for the ‘amazon’ Category

S3 outage – We weren't affected

February 15, 2008 68 comments

Amazon S3 had an outage today. First I knew about it was reporters emailing and calling me asking if we were knocked out by it.

We weren’t. No customers reported issues, and our systems were all showing typically low and acceptable error rates. To be honest, I’m surprised.

I wasn’t going to blog about it until I understood why we weren’t affected, but I’m really getting inundated with requests now, so I figured this would be a good way to optimize my time rather than spending all day on the phone. 🙂

We’re researching what happened now, but again, I didn’t know about the outage until after it was over, and I haven’t spoken to anyone at Amazon yet. Until I finish my research and speak with Amazon, I’m not going to speculate on what may have happened or why.

I can say, once again, that we pay the same rates everyone else pays and that, other than some early access to upcoming beta services, we don’t get any preferential treatment that I’m aware of.

Some thoughts, though:

  • We expect Amazon to have outages. No website I’m aware of doesn’t, whether it’s Google, Amazon, your bank, or SmugMug.
  • I’ve written about Amazon S3 outages in the past, but in the last ~12 months, we’ve only seen a single ~2 minute outage window (on January 22nd, 2008 at around 4:38pm Pacific). We also had one recent fairly major hiccup with EC2.
  • Yes, I believe there will probably be times where SmugMug is seriously affected, possibly even offline completely, because Amazon (or some other web services provider) is having problems. Today wasn’t that day. Nobody likes outages, especially not us, but we’ve decided the tradeoffs are worth it. You should have your eyes wide open when you make the decision to use AWS or any other external service, though. It will fail from time to time.
  • We’ve done our best to build SmugMug in such a way that we handle failures as gracefully as possible. We can’t cover every case, but I think that the fact that we didn’t experience customer-facing outages today is a testament to that. Again, I want to stress that we do expect Amazon to cause us (rare) outages in the future, and that’s unavoidable, but today we dodged that bullet.
  • Amazon’s communication about this has been terrible. It took far too long to acknowledge the problem. Fixing a major problem can take forever, which is understandable, but communicating with your customers should happen very rapidly. Amazon’s culture, internally, is very customer focused, so this is a strange anomaly. I will definitely lean on them some about it, and everyone who was affected should rightfully howl too.
  • I’ve asked Amazon repeatedly for an “Amazon Web Services Health” page that shows the current expected state of all their services. Then you can tell at a glance (and even poll and work into your own monitoring) whether any of the services are having problems. Something like Keynote’s Internet Health Report would be a good start, but as Jesse Robbins points out, is the gold standard. This page could also double as a mechanism to let customers know what’s being worked on and current ETAs when there are problems.

I’ll try to post a follow-up about why we weren’t affected when I know more. It’s possible that some of the reasons we survived was due to some of our “secret sauce” and I just won’t be able to say, but I kinda doubt it.

Bottom line: While the outage was certainly a big deal to those affected, I think the bigger deal here is how Amazon handled the outage. They need to communicate better about these mission critical services and their health.

If I didn’t answer any questions you’d like me to answer, please post a comment and/or send me an email. I’ll do my best to respond.

UPDATE 1: I’m not sure why there’s all this confusion, but SmugMug *does* use Amazon as our primary data store. We maintain a small “hot cache” in our datacenters of frequently/recently viewed photos and videos, but there are massive numbers of them that are only at Amazon. This is a change from our initial usage of S3, and the change is based on how reliable they’ve been. Yes, we still consider them to be very reliable even after an outage like this. And yes, I suspect our “hot cache” did at least partially enable us to ride out this issue.

Categories: amazon Tags: , , , , , , ,

Amazon DevPay is out (in Beta)!

December 19, 2007 3 comments

They’re coming fast-and-furious now. 🙂 First SimpleDB, now DevPay.

DevPay basically lets you layer your own service offerings on top of Amazon’s and get compensated for doing so. I can’t tell you why we’re so excited about this one, just yet, but we have some neat ideas we’re playing with that I hope I can talk about soon. Meanwhile, if you’re a developer wanting to re-sell Amazon’s services with your own special sauce mixed in, now’s your chance. 🙂

This service is particularly interesting, to me, because it doesn’t stand on its own. Most of the other services work great together, but you can easily use them independently to do interesting things. I’ve been expecting layers of services to start forming where some of the core components are pre-requisite building blocks (say a CDN on top of S3 or a Load Balancing API on top of EC2), and I think DevPay is the first of these. Can’t wait to see what kind of neat businesses come out of this.

Categories: amazon

Amazon announces SimpleDB (in Beta)

December 14, 2007 7 comments

Sweet! Amazon finally took the wraps off of SimpleDB. They’ve been working on this for awhile, and as you can probably tell, it’s a natural fit with S3 and EC2. There’s a great write-up about it over on inside looking out.

This is nearly a perfect solution for some of our data-related scaling challenges, except for two issues:

  • Physical proximity. Some of my datacenters aren’t close to Amazon’s, so the actual time to query SimpleDB is query time plus latency. This isn’t a problem if you’re doing all your queries from EC2, but we’re not there yet (we’d like to be, but a few pieces are missing. SimpleDB is one of those pieces, so we’re getting closer…). Amazon has promised me they’re workin the speed of light issue. 😉
  • Attribute size limits. We have some data fields that are longer than 1024 bytes (most aren’t and would work fine). We’ve thought about chunking the data up to get around this, which is a possibility, but it gets messy. Storing them in S3 is both overkill and probably too slow – if I need to get a few thousand photo captions *fast*, doing it through S3 isn’t optimal. If we could solve the latency problem I already mentioned, I’d be fine storing that specific data in some other store and working around it that way.

On the plus side, SimpleDB should be screaming fast, incredibly scalable, and almost all of our SQL queries would work with no changes other than syntax. Like many of you, I’m sure, we’re using much of our RDBMS as a fairly simple data store and aren’t using many advanced RDBMS capabilities. All of those queries could just use SimpleDB and then we could devote our DB iron to just the rare complex queries. We’re not alone – tons of web apps are gonna love this.

I’m thrilled to see the Amazon AWS stack continue to grow, and I’m shocked that they have as big of a lead as they do. I would have thought Microsoft / Google / Sun / whomever would have been out with some competition by now. It’s gonna happen – but I never would have guessed it would take this long.

Oh, and while I have your attention – SmugMug is now a fairly heavy user of EC2 and I have a write-up coming. So check back later if that’s of interest.

Categories: amazon, datacenter

Kindle Review

November 20, 2007 37 comments

Got my Kindle today and have been playing with it for a little while. FYI, I’ve owned a Sony Reader since the day it shipped, too. I’ve always loved to read, and as a result, I’m usually reading multiple books at once. I think I’m currently partway through at least 10 of them, so when I travel, it’s hard to know what kind of mood I’ll be in and I have to pack lots. And they’re heavy. A device like the Kindle or Reader is the holy grail for someone like me, plus I’m a gadget geek, so I’m always an early adopter.

I’ve loved my Sony Reader, the device itself, since I bought it. But the Sony Connect software is so unbelievably bad, I can’t really bring myself to use it. Combined with a terrible selection of books, the Reader has been mostly worthless, despite being a fairly great device. Which is really sad. When I heard the Kindle had the ability to just buy books with a click, no software required, I knew I had to at least give it a try.

Here’s a brief rundown of my thoughts so far:

  • It’s not ugly. I’ve been moaning to everyone who would listen that it’s an ugly device for awhile now, but in person, it’s actually small, light, and understated. It’s certainly not an Apple-designed wonder, but it’s hardly ugly, either. And I find I enjoy reading on the white device (Kindle) more than the black one (Reader), probably due to the contrast or maybe just the white gets “out of the way” more easily, letting me slip into the book.
  • The display is gorgeous, just like the Reader. I have zero problems with it. I’m already used to the lag when you change pages from using the Reader, and I found that after just a few days of using the Reader, I was already hitting the Next Page button early enough so the page turned just when I wanted it to.
  • The scroll thing is very cool and fun to use. I got some ‘oohs’ and ‘aahs’ here at the office as worked it’s magic.
  • The Next/Prev page buttons are much nicer than the ones on my Reader.
  • The EVDO works like magic. Buying books online with a click, no software required, is the killer feature here, and it’s perfect.
  • It comes pre-associated to your Amazon account, so there’s no configuration. Just pick it up, turn it on, and start browsing & buying. *Very* cool.
  • The book selection sucks. There are big gaps, even for well-known bestselling books. Having worked in the book industry before, I put most of this blame in publisher’s laps. They’re just a nightmare to deal with, and paranoid about their content. Apparently they don’t want my money or yours, and even Amazon doesn’t have the weight to make them see reason. Shades of the music, TV, and movie industries, anyone? This must be incredibly frustrating to Jeff and everyone else at Amazon.
  • The web browser worked surprisingly well in the few minutes I used it. But at, I was getting some strange results. Using my Mac’s browser, I searched for a product and quickly found it. But on the Kindle, it failed to turn up the same results, even when I searched for very specific words in the title and description. I was hoping to buy it from my Kindle, but alas, I had to do it the old fashioned way.

More about the book selection… Wanting to have a reference library on my Kindle, I started looking for some of my favorites I refer to often. I searched for Good to Great, my all time favorite business book, first and instead managed to find a weighty tome of eight papal social encyclicals that I could buy for the low low price of $100.80. Built to Last isn’t there either. Pour Your Heart Into It? Nope. Innovator’s Dilemma? Nope. They did have The Wisdom of Crowds and The Culture Code, both of which I bought.

On to a personal favorite category of mine, historical fiction, I knew they’d have James Clavell’s Shogun – but it wasn’t to be. Nor most of Colleen McCullough’s awesome Masters of Rome series, though luckily The October Horse is available (and now on my Kindle). Fine, how about history – Guns, Germs & Steel. Bzzt, no joy. How about the best book ever written on health & nutrition – The China Study? Nope, none of that either. (They did have another excellent book, the Omnivore’s Dilemma). Luckily they did have my favorite fantasy author of all time, George R.R. Martin, and his Song of Ice and Fire books.

There were some strange ones, too. For example, they do have The Daring Book for Girls – but not The Dangerous Book for Boys. And you would be surprised at how many of my search terms turned up weighty, expensive scholarly works rather than the books and novels I was looking for.

Bottom line: I love the device, I love the Kindle Store interface, I just wish the selection was better. I believe it’ll be a permanent resident in my laptop bag, going everywhere I go, but it looks like I’ll still be forced to lug around some dead trees. 😦

UPDATE: Spent quite a bit more time with it last night doing what it’s really supposed to be used for – reading – instead of playing with all the whiz-bang features. I curled up on the couch with Jedi and the Kindle and got lost in The October Horse. And I’ve gotta tell you – it’s a much better device than the Sony Reader in lots of subtle ways:

  • The Next Page buttons are in *much* better places and are much easier to use than the Reader. I tend to curl up on my side to read a lot, switching sides every 15-30 minutes, and the Kindle is easy and natural to hold in such a way that turning to the next page is very easy.
  • I was right about the white color – it makes it easier to sink into reading as well as read the text. The device does almost get out of the way better than the Reader because it’s white.
  • The flicker when you turn pages is still mildly annoying, just like it is on the Reader, and distracts you from completely sinking into the book… at first. Just like the Reader, though, a couple of pages in, your brain has magically figured out the right timing to turn the pages at precisely the right time so the flicker happens as your eyes are traveling from the bottom-right to the top-left of the page. Then the flicker feels no more jarring than turning a page on a dead tree.
  • The cover sucks for reading. Seems like it’s probably great for when I throw it in my laptop bag, but the Sony Reader’s cover is a much better fit, less bulky, etc. I guess the Kindle’s cover needs to be sorta bulky so it can prop the book up for optimal EVDO range, but it still sucks. Take it off to read – it just gets in the way.
  • The text, just like the Reader, is just as easy to read as any book. I have the font on the smallest setting, to minimize page turns, and it’s beautiful and legible. That part of the experience is literally almost indistinguishable from reading a book.

More as I get it… 🙂

Categories: amazon, personal

I get SLAs now. Duh.

October 11, 2007 5 comments

Ok, so I guess I’m a total n00b. In hindsight, SLAs make a lot of sense after all. The whole point isn’t to compensate SmugMug for our loss, it’s to make it unprofitable for the service provider to keep making the same mistakes.

In other words, let’s say Amazon’s margins on S3 are 15%. (I have no data, I’m just picking that number out of the air). If Amazon has a serious problem during a month, they have to cough up 25% to all their customers. In other words, they lose 10% instead of make 15%.

That’s pretty major incentive – and it now totally makes sense why SLAs are so highly valued.

Carry on.

Categories: amazon, business, datacenter

Amazon S3 SLA is here! (Nirvanix dies?)

October 8, 2007 17 comments

Amazon has finally released and put into effect their SLA for S3. I know a lot of my readers will be thrilled about this. 🙂

I’ve gotten a few questions about Nirvanix in the past month or so, especially about the fact that they offer an SLA (and that S3 didn’t). I think this probably puts the final nail in Nirvanix’ coffin because:

  • Why would you trust Nirvanix, a no-name company, with your precious data?
  • Worse, they’re affiliated with MediaMax/Streamload in some way, who have a reputation of poor service. (I’ve even seen reports of data loss at Streamload, though I haven’t bothered to check).
  • Just how much is an SLA worth when there’s nothing behind it to back it up?
  • They’re more expensive than Amazon. Um, duh.

SLAs don’t mean a lot to us, anyway, as I’ve said before because:

  • Everything fails sometimes.
  • The SLA payment is rarely comparable to the pain and suffering your customers had to deal with.

But I know it’s very important to lots of people, so I expect there’s cheering and dancing in the streets. 🙂

UPDATE: I get SLAs now. Sorry for being dumb.

Categories: amazon, datacenter

Speaking at 'The Startup Project' Wednesday

September 11, 2007 2 comments

I should have posted this awhile ago. I suck. I’m sorry.

Anyway, I’m speaking at The Startup Project, an Amazon and Kleiner-Perkins event tomorrow in the Silicon Valley. I’ll be talking a little bit about S3, EC2, and FPS, the three announced Amazon Web Services we’re most excited about.

There’ll be a Q&A, and I’m happy to stick around after and answer questions about AWS or anything else under the sun, too, if you have any.

See ya there! RSVPs are required, I believe.

(On a related note, I blew it this year and spoke at and attended too many events. In 2008, I’ll be going to far fewer conferences and will be very selective of the ones I speak at, so if you think I’d be a good fit with your event, ask earlier rather than later please)

Categories: amazon, smugmug
%d bloggers like this: