Amazon S3: Show me the money
UPDATE 4/30/07: This post was written in November 2006, so these numbers are a little out of date. It’s now been 12 months and we’ve saved almost exactly $1M. You can see the most recent numbers, as of April 2007, in my ETech slides.
I still have some more Web 2.0 Summit stuff to write up if I get a few minutes today, but let me talk about Amazon’s S3 for a minute. At the conference, I was chatting with Michael Arrington of TechCrunch fame (who perfectly handled a blogosphere mini-explosion last week, I thought) and we got to talking about S3. He was impressed with how we were using it, but joked that our $500K saved number sounded like “complete bullsh*t”. I laughed along with him and assured him it was true, but on the way home I got to thinking that it IS a really big number to throw out there without details.
So here are the cold hard facts:
- Our estimate, as you can see in BusinessWeek’s cover story, is that we’re saving $500K per year. We’ve been using S3 for almost 7 months so far (we launched it on or around April 14th), so for my $500K estimate to be in the right ballpark, we should be somewhere near $291K saved to date (well, we don’t grow linearly, so less than that … but let’s do easy math, shall we?).
- We had roughly 64,000,000 photos when we launched S3. We now have close to 110,000,000 photos. Yes, that’s ~72% growth in 7 months.
- To sustain our pre-S3 growth, we were buying roughly $40,000 per month in hard disks plus servers to attach them to. We’re not talking about EMC or other over-priced storage solutions. We’re talking about single processor commodity Pentium 4 servers attached to really cheap Apple Xserve RAID arrays. Not quite off-the-shelf IDE disks, but once you factor in the reliability and managability, the TCO comes out to be in a similar ballpark (We’ve done it both ways).
- If you’re doing the math at home, $40K may seem a little high until you realize how our architecture works: We use RAID-5, with hot spares, and we have two entirely separate storage clusters. That means we have to buy 1.4TB of raw disk to store an actual 500GB.
- To sustain our current, Nov 2006 growth rate, we’d need to buy more like ~$80K per month. Let’s assume over the 7 months, it ramped from $40K to $80K linearly (it was actually more of a curve, but this makes the math easier). $40K + $46K + $53K + $60K + $66K + $73K + $80K = $418K
- Our datacenter space, power and cooling costs for those arrays is ~$1.36/month for every $100 of storage. (~$544month @ $40K, ramping to ~$1088/month @ $80K). $544 + $626 + $721 + $816 + $898 + $993 + $1088 = $5,686.
- It’s cost us some manpower to move everything up to S3. So while I expect to save money on manpower in the long run, currently it’s probably break even - I don’t have to install, manage and maintain new hardware, but I’ve had to copy more than 100TB up to Amazon. (We’re not done copying old data up yet, either)
- Total amount NOT spent over the last 7 months: $423,686
- Total amount spent on S3: $84,255.25
- Total savings: $339,430.75
- That works out to $48,490 / month, which is $581,881 per year. Remember, though, our rate of growth is high, so over the remaining 5 months, the monthly savings will be even greater.
- These are real, hard numbers after using S3 for 7 months, not our projections. They closely match (but are actually slightly better) than our projections.
So there you have it.
But wait! It gets even better! Because of the stupid way the tax law operates in this country, I would actually have to pay taxes on the $423K I spent buying drives (yes, exactly like the money I spent was actually profit. Dumb.). So I’d have to pay an additional ~$135K in taxes. Technically, I’d get that back over the next 5 years, so I didn’t want to include it as “savings” but as you can imagine, the cash flow implications are huge. In a very real sense, the actual cash I conserved so far is about $474,000.
But wait! It gets even better! Amazon has been so reliable over the last 7 months (considerably more reliable than our own internal storage, which I consider to be quite reliable), that just last week we made S3 an even more fundamental part of our storage architecture. I’ll save the details for a future post, but the bottom line is that we’re actually going to start selling up to 90% of our hard drives on eBay or something. So costs I had previously assumed were sunk are actually about to be recouped. We should get many hundreds of thousands of dollars back in cash.
I expect our savings from Amazon S3 to be well over $1M in 2007, maybe as high as $2M.
Perhaps most important, though, is the difficult-to-quantify time, effort, and mental thought we’re saving. We get to spend both that money and all of our extra time and effort on providing a better customer experience and delivering better customer service. Storage was a necessary evil that’s now been nearly removed as a concern.
Want more? I have some other posts on the subject:
And I’ll continue to post with more hard details, including our technical architecture and some of our code, as well. And yes, we’re starting to consume other Amazon services like EC2.
November 10th, 2006 at 10:39 am
Thanks for the write up, and I would love to see more technical details of how you manage your S3 system. I have a few questions…
I gathered from your previous posts that S3 was just going to be your backup storage, but it now sounds like you are going to put all your eggs in one S3 basket? Whats your backup/alternative when S3 breaks or doesnt meet your needs for whatever reason? Have you thought of the legal ramifications of using S3, like when Amazon stores your data in other countries?
November 10th, 2006 at 10:59 am
Hi D,
Over the last 7 months, we’ve experimented with Amazon S3 in a variety of roles. Initially, and for much of the time, they’ve been relegated to backup and disaster recovery only.
That’s definitely changed. With a 7 month track record of being more reliable than our own storage, the tables have turned a bit. S3 is now our super-reliable piece of infrastructure, and our own storage is beginning to serve another purpose.
I’ll explain all of this in detail in some future blog post when I get some time, but we’re not exactly putting all of our eggs in one basket, although we’re at a point that I think it’d be possible and feasible to do so.
Don
November 10th, 2006 at 11:10 am
[...] Interesting writeup by Don Macaskill, CEO of SmugMug, who tells why Amazin is saving his photosharing service big bucks. Filed under: blogging @ 11:10 am # [...]
November 10th, 2006 at 12:00 pm
I think it could be risky to put your eggs in other company, which can be reliable, but how do you know they won’t quit the service? How much time they guarantee it will be running? I suppose you thought about this, but I didn’t see the answer.
Thanks for these great posts. Very pleased to know about these things from you, Smugmug
November 10th, 2006 at 1:16 pm
There are no guarantees in this business. My hard drives fail in my datacenter, so I have to work around that. A larger company bought up all the power at one of our datacenters, so we had to move or die. The list goes on and on.
Jeff Bezos is behind this 100%, I’ve spoken to him personally about it a few times, and I believe they’re not going to “quit the service”. But there are no guarantees. If they do decide this isn’t for them (I really, really doubt it because it makes perfect sense for them to do this, and I think they’ll make money), we’ll come up with something new.
Don
November 10th, 2006 at 2:07 pm
[...] MacAskill » Blog Archive » Amazon S3: Show me the money | del.icio.us | Digg it | Furl | reddit | StumbleUpon| [...]
November 10th, 2006 at 2:20 pm
Thanks for that write up. Nice to see how smugmug is able to use the services of other companies to improve its overall business and bottom line.
“I’ll save the details for a future post, but the bottom line is that we’re actually going to start selling up to 90% of our hard drives on eBay or something”
Would you consider offering current smugmug users a “private” auction of some type (if it’s feasible)?
November 10th, 2006 at 2:28 pm
So, if we buy the drives on eBay, do we get all the photos on that drive as well?
November 10th, 2006 at 2:35 pm
Selling off hard drives….. nice idea. I want some.
November 10th, 2006 at 2:49 pm
I’ll definitely post on my blog when we’re ready to start selling the storage, for those who are interested. SmugMug customers will get first preference.
November 10th, 2006 at 2:50 pm
Don please let me know how my figures are faulty:
Let assume it costs $1000 to store 500GB.
The $1000 is a one time cost. If you go with S3 you will pay 500 x $.15 = $75 per month. In 2 years with S3 your cost will be 24months x $75=$1800.
$1000 vs $1800 and counting
So how does S3 save you money over the long run.
Sure my numbers are off. It may cost $2000+ to store 500GB, going with S3 you will be paying a reoccurring fee instead of a 1 time when you purchase your own equipment. Sorry it just seems like Im missing something big here.
Thank you
November 10th, 2006 at 3:05 pm
Part of Amazon’s S3 license agreement is concerning, section B4 where it says:
“(ii) Your S3 Content is not obscene, libelous, defamatory or otherwise malicious or harmful to any person or entity;”
That seems incredibly broad, as obscenity is “in the eye of the beholder”, and there’s a group out there to protest just about any type of content out there (porn, of course, but then there are the anti-video game nuts, and even some that believe photography steals souls (ok maybe they’re all gone by now)). Is that just Amazon’s way of saying: “We can shut off your business at any time, if we get a complaint”?
November 10th, 2006 at 3:57 pm
You’ve been promising more details since your first post in August, lol. Which is how I found your blog to begin with. You’d be doing a lot of people favors by posting that and helping speed up the adoption of S3 I think too.
We also use S3 for a significant amount file hosting for our customers and have found it to be more reliable than our own. Our savings aren’t as dramatic because the files are used more frequently and we eat up much of the savings in transfer costs. Our bandwidth is about 6.5x cheaper than what Amazon charges and we do a lot of it. Peoples personal photos aren’t view all that frequently so the amount of drive space you consume is large compared to the amount data transferred so you’re savings are more substantial.
I’m looking forward to you technical article. We’re looking for ways to do things better and cheaper and we’re doing a pretty big scale so there isn’t much real world info on companies our size using S3. I was hoping the article would validate some of the ways we were looking to change our implementation to make S3 a little bigger piece of our solution. I’m sure we could be saving another $100k a year or more.
We struggle with the same issues, such as whether to use it as a primary or just as a backup as we do now, or to some degree in between. What if Amazon quits? What if they fail for even a few hours, 5,6,10. Etc.
Our data is a slightly more time sensitive and needs high availablity. On the other hand it costs a lot more money to store the info ourselves. Because our data is accessed more frequently the Apple RaidX servers aren’t an option. We use 15k scsi drives on generic high end computers. It runs about $20k per 1.6T of storage. So the storage savings we see are more significant than SmugMug, but the bandwidth savings for us are far less.
I would agree though the savings overall between thought power, time, server purchases, reliabilty gains, and so on are pretty big. They would be in the low-mid 6 digits for us as well and already are in the low 6 digits.
November 10th, 2006 at 4:49 pm
Way to go! Im sure your whole purpose of writing this little blog was to help others!!!! Nah, either you or someone gave you the idea to use this supposedly helpful post to whore your business via Digg.com.
Congrats! It worked…I now have heard of smugmug.com and possibly a quarter of million others too. Marketing at it’s best…….cheap!
Peace,
Chaser
November 10th, 2006 at 4:50 pm
90% disk savings? Sounds like you’re planning to stop storing originals at your data center and just fetch them from S3 when you need them. Latency is a lot less of a concern at that size.
November 10th, 2006 at 5:34 pm
[...] Update: Smugmug explains how using Amazon S3 saves them $500,000 a year by curiouscat Tags: Management, Innovation, Creativity Permalink to: Amazon Innovation [...]
November 10th, 2006 at 6:19 pm
I have been looking into EC2 as well and the only drawback that I can see is the lack of persistent storage.
Have you figured out a way around this limitation or does it even matter to you?
November 10th, 2006 at 7:00 pm
@John:
Actually, you’re off base on the cost basis. It doesn’t cost us $1000 for 500GB - it’s closer to $3000 for just the disks and servers.
At $75 per month, that’s 40 months. But that doesn’t take into account the datacenter space, the cooling and power costs, the human labor to install, monitor and repair the hardware, and replacment costs of drives and other hardware, which do fail.
Perhaps most importantly, though, you’re forgetting that over time, Amazon’s costs will drop as the price of storage drops. That’s *not* true of disks I buy - those are sunk costs.
Hope that helps!
Don
November 10th, 2006 at 7:01 pm
Now I am a bit concerned. I am a Smugmug customer and was not aware until now that my data was being stored at Amazon.
Do you have my permission to do this ?
How secure is my data ?
What is Amazon’s promise regarding the protection of my data ?
Are you still storing my data in three separate data centers as advertised ?
Can I opt out of Amazon having my data ?
Regards,
Ross
November 10th, 2006 at 7:04 pm
@Chaser:
I know you’re just trolling, but I’ll take the bait.
SmugMug is a popular, profitable, growing business. Of course I like having a few thousand diggers read my blog - who wouldn’t? But it’s not exactly earth-shaking marketing for us.
We’re in the cover story of BusinessWeek this week. Walt Mossberg of the Wall Street Journal recommended us two weeks ago. PC Magazine has given us the Editor’s Choice award year after year.
As much as I love digg personally, it’s not exactly the same caliber.
I wrote this post because, as a geek, I love the thought of not having to worry about these things as much - let Amazon do it. If people enjoy the post, great. If not, big deal.
Welcome to blogging.
November 10th, 2006 at 7:06 pm
@Tracy:
S3 provides all the permanent storage you’re likely to need, unless you’re doing really really heavy IO stuff, like a heavy database.
Our EC2 use-case doesn’t rely on anything except S3 for storage, so we’re set.
November 10th, 2006 at 7:09 pm
@Ross:
We’re storing your data in multiple datacenters, yes. Most are Amazon, some are SmugMug.
We don’t own any datacenters ourselves, so your data has always been stored “at” some other company, who could theoretically grab it. Of course, that would kill their business, and we have reasonable assurances that our various providers, including Amazon, won’t do that.
Your data is private and secure at Amazon. No-one but SmugMug has access to it. It’s your data, managed by SmugMug, and Amazon happens to be the brand of disks we’re using, essentially.
We’re happy to answer specific details or questions - just email our Support Heroes here at SmugMug.
November 10th, 2006 at 10:32 pm
Onethumb,
How are you finding EC2?
November 10th, 2006 at 11:18 pm
I’m curious if there is a service similar to S3 for MySQL or any other database platform. Or does S3 allow MySQL queries?
I have a MySQL database nearing 100 terabytes, and still growing. I’d love to sell of my hard disks and migrate to a service that can manage it.
November 11th, 2006 at 1:51 am
What about long-term costs ?
Like 3-5 years running on own HDDs vs. Amazon ?
With HDD’s you have to pay for electricity only - but with Amazon more.
As well - what will happen if Amazon will decide to increase their prices(after reading your blog
- how much it will cost you to migrate to another datacenter or get your own ?
November 11th, 2006 at 3:16 am
Initially i was going to say well if you’re saving money by using S3 how does S3 manage to store all that data at a reasonable cost - but there is a lot there we don’t know - for example the savings they are making by the massive volume they represent…
but then it occured to me that if S3 should fail at any stage I guess you could roll your savings back into a storage system which by then would be much more storage for “todays” money.
I confess I love the idea of S3 and more so the Idea ofS3 combined with Elastic Cloud (way cool name btw =)) .
November 11th, 2006 at 7:57 am
@TAG:
The long-term costs are better with Amazon because their prices fall in line with HDDs, whereas my initial purchase price of HDDs is fixed and sunk.
With our projections, if Amazon’s costs don’t fall (they will, but as an exercise we projected both), Amazon starts to cost more about 9 years out. If they do fall reasonably well with the market cost of HDDs, Amazon never costs more. Not after decades.
November 11th, 2006 at 9:10 am
[...] Amazon’s S3 service (I need to look into this) offers unlimited data storage and transfer at low flat rates, enabling start-ups or more established companies to focus on building their business and traffic, instead of how to scale their server space. Don MacAskill, CEO of SmugMug, details how S3 has saved his company well over $500,000 in the last seven months, and how he expects savings of well over $1M in 2007. He was spending that money somewhere else before he made the change to S3, so for whoever those vendors were, some warm water is coming under their iceberg. [...]
November 11th, 2006 at 10:45 am
You talk about the costs of storage but I haven’t seen you address the bandwidth costs. I’d be curious to see your data on the $0.20 per GB of data transferred. This is a more subjective debate, since bandwidth usage is going to vary by company. (some companies will have usage fairly distributed if they have a global audience, while others serving mainly the US will have spikes and lows)
I’m sure it’s nice to have the high burst availability of someone like Amazon, but I would think that since the costs are linear, it might kill you.
Thoughts?
–Ed
November 11th, 2006 at 12:03 pm
@Ed:
Our bandwidth costs with Amazon are higher than our initial projections because we’ve relied on them more than we thought we would (they’ve been more reliable than our own internal storage).
Their cost of bandwidth, if you buy bandwidth in large chunks like we do, is high. We don’t pay any of our providers nearly as much as we pay Amazon, but we buy bandwidth in GigE connections, so we get pretty good rates.
The honest answer, though, is that our bandwidth charges to Amazon have been “lost in the noise”. We’re saving so much on overall storage, that I don’t terribly mind the extra bandwidth expense - it’s certainly not enormous.
That being said, we run our business as efficiently as we can (we have no investment, no debt, and are profitable … keeps you on your toes), so we’re actually going to be buying / leasing GigE connections to Amazon’s datacenters and peering directly with them to reduce the $0.20/GB cost and to eliminate the costs associated with our bandwidth providers for transfers to/from Amazon.
November 11th, 2006 at 12:16 pm
The $500K didn’t seem like such a big number until I realized this was an 80% reduction in costs … WOW!
November 11th, 2006 at 2:49 pm
[...] I have been talking about how the Amazon’s web services like S3 and EC2 are helping the startups cut down on their costs. The CEO of Smugmug, a photo sharing site, writes a blog post on how his company saved $500 K with S3. He, then, goes on to predict that they will save upto $ 2 Million in the next year. I am pretty convinced that Amazon Web Services like S3 and EC2 are going to change the tech playing field in a huge way. [...]
November 11th, 2006 at 5:25 pm
I have no expertise in the area, but using Xserver RAID sounds like a premium price when there must bigger, better players when you are talking the quantity and quality of storage that you are.
November 12th, 2006 at 10:36 am
[...] SmugMug, a photohosting company has a blog where the CEO writes about some real savings that are happening right now: We’ve been using S3 for almost 7 months so far (we launched it on or around April 14th), so for my $500K estimate to be in the right ballpark, we should be somewhere near $291K saved to date (well, we don’t grow linearly, so less than that … but let’s do easy math, shall we?). [...]
November 12th, 2006 at 12:13 pm
[...] Michael Arrington called them on their numbers during Web2.0 Summit, and so the Smugmuggers have come back with some real numbers on their blog. The conclusions are: [...]
November 12th, 2006 at 1:54 pm
@Lloyd:
You’d be surprised. Everyone from Oracle to eBay use Xserve RAID now, and use it heavily.
Apple went from selling no enterprise storage to being the 11th largest seller on the planet in just a few years.
Their cost per GB is extremely competitive, orders of magnitude better than any other major player. The only reasonable cheaper alternative I know of is to use commodity disks and servers - but that dramatically increases your labor costs, something I’m averse to.
November 12th, 2006 at 3:39 pm
[...] SmugSmug e’ un sistema per condivisione di foto on-line che si appoggia ad S3. Hanno pubblicato i loro calcoli, reali e concreti, su quanto abbiano risparmiato ad oggi optando per S3 come storage on line. Tanto per gradire: In a very real sense, the actual cash I conserved so far is about $474,000. [...]
November 13th, 2006 at 1:31 am
[...] L’altro è puramente tecnico: dal blog di SmugSmug, commenti sull’uso di Amazon S3 per mettere su la propria startup web e come risparmiare montagne di soldi. [...]
November 13th, 2006 at 7:36 am
Hi onethumb.
Although I appreciate the transparancy of the business hoodahaada, I’m alot more interested in what you’re going to spend these savings on. You briefly mention customer experience and support, could you elaborate on this?
Malte
November 14th, 2006 at 6:24 am
[...] This was also the subject of a recent cover story on Business Week (written by Rob Hof, one of Silicon Valley’s best tech writers). While these services are still in their infancy, a high profile group of companies are starting to jump on board: Microsoft, Xerox and Second Life are all customers. And SmugMug CEO Don MacAskill recently wrote a long blog post gushing about the amount of money they are saving using S3 over alternative solutions - $500k so far and an estimated $1 million or more in 2007. [...]
November 14th, 2006 at 10:33 am
S3 seems to be a good service from amazon. SQS seems to be another good service from their offering
November 14th, 2006 at 10:36 am
@Malte:
We’re already rolled out new features and hired new employees that we wouldn’t have otherwise been able to do, so it’s not like we’re building up all this cash so we can hopefully figure it out.
The chunk of Pro features we released a few months ago was a direct result, as was the recent hiring of a number of new Support Heroes to handle customer service.
Hope that helps!
November 14th, 2006 at 10:38 am
@dhoom:
Yeah, SQS is very cool. Unfortunately, for a company our size, the cost per transaction seems to be a little too high, particularly when we already have internal queueing services we can just expose. But for a startup, it’s awesome!
November 15th, 2006 at 4:59 am
[...] C’était aussi le sujet d’un article récent dans Business Week (écrit par Rob Hof, un des meilleurs auteurs technologiques de la Silicon Valley). Tandis que ces services sont toujours à leur démarrage, un groupe de grosse de sociétés commence à sauter à bord : Microsoft, Xerox et Second Life en sont tous des clients. Et le PDG SmugMug Don MacAskill a récemment écrit un long billet à propos de l’économie réalisée en utilisant S3 sur d’autres solutions alternatives - 500K$ jusqu’ici et 1 million de $ ou plus évalué pour 2007. [...]
November 15th, 2006 at 1:05 pm
[...] The benefits for all of this are obvious I think. You would have every startup beating on your door to host their solutions. It would save fund-strapped companies tens or hundreds of thousands of dollars in hardware costs. Need proof? Go read this post by the SmugMug CEO, who is using Amazon’s storage solution. He has saved several hundred thousand in less than a year in hardware costs. I’ll bet there are a wealth of startup ideas out there that haven’t passed the idea stage because of the daunting prospect of needing the funds for hardware alone. I personally am one of those individuals by the way. A co-worker and I recently did a comparison of what our company spent for hardware for a relatively simple web application solution. They spent approximately $20,000 in hardware. That may not be much, but if they went with something like what Amazon is offering at their prices, that 20k would get spread out over 13 years. 13 YEARS! [...]
November 15th, 2006 at 1:16 pm
[...] This blog post by SmugMug’s CEO on how he thinks using Amazon’s Simple Storeage Service to run the photo-sharing site is saving him $500,000 USD / year. See also his post Amazon + Two Guys + $0 = Next YouTube. See also Web Services Success Stories at Amazon. [...]
November 15th, 2006 at 9:27 pm
[...] SmugBlog: Don MacAskill » Blog Archive » Amazon S3: Show me the money “I thought) and we got to talking about S3. He was impressed with how we were using it, but joked that our $500K saved number sounded like “complete bullsh*t”. I laughed along with him and assured him it was true, but on the way home I got to thinking (tags: startup economics storage WebServices Amazon s3 smugmug) Tags: Leave a comment [...]
November 17th, 2006 at 1:23 pm
[...] SmugBlog: Don MacAskill » Blog Archive » Amazon S3: Show me the money Our estimate, as you can see in BusinessWeek’s cover story, is that we’re saving $500K per year. We’ve been using S3 for almost 7 months so far (we launched it on or around April 14th), so for my $500K estimate to be in the right ballpark, we should (tags: s3) [...]
November 21st, 2006 at 3:55 am
[...] This was also the subject of a recent cover story on Business Week (written by Rob Hof, one of Silicon Valley’s best tech writers). While these services are still in their infancy, a high profile group of companies are starting to jump on board: Microsoft, Xerox and Second Life are all customers. And SmugMug CEO Don MacAskill recently wrote a long blog post gushing about the amount of money they are saving using S3 over alternative solutions - $500k so far and an estimated $1 million or more in 2007. [...]
November 28th, 2006 at 2:19 am
[...] I’ve been looking at SmugMug for offsite photo storage based on some nice things I’ve been reading about them and some conversations in the office. They use Amazon’s S3 for very cost-effective storage which, being a geek, made me think that perhaps I could simply load my photos into S3 for storage. Yes of course, but then I’d lose the neat management tools and features that SmugMug provide. So my thought is: when are we going to see applications that let me use my own S3 account rather than theirs? [...]
December 1st, 2006 at 12:40 pm
[...] Amazon’s web services unit (see our discussion of Mechanical Turk, Simple Storage Service (S3) and Elastic Compute Cloud (EC2)) has been making waves this year, and a number of high profile startups are using one or more services to speed deployment time and save money. In a recent interview, Amazon CEO Jeff Bezos discusses Amazon’s strategy around their Web Services unit. [...]
December 1st, 2006 at 12:58 pm
[...] Amazon’s web services unit (see our discussion of Mechanical Turk, Simple Storage Service (S3) and Elastic Compute Cloud (EC2)) has been making waves this year, and a number of high profile startups are using one or more services to speed deployment time and save money. In a recent interview, Amazon CEO Jeff Bezos discusses Amazon’s strategy around their Web Services unit. [...]
December 1st, 2006 at 1:16 pm
[...] Amazon’s web services unit (see our discussion of Mechanical Turk, Simple Storage Service (S3) and Elastic Compute Cloud (EC2)) has been making waves this year, and a number of high profile startups are using one or more services to speed deployment time and save money. In a recent interview, Amazon CEO Jeff Bezos discusses Amazon’s strategy around their Web Services unit. [...]
December 1st, 2006 at 3:44 pm
What do you mean by this part:
Why would you have to pay taxes on your purchases as if they were profits? You should be able to depreciate that expenditure, not pay tax and then get it back.
December 1st, 2006 at 3:48 pm
The tax discussion you mention doesn’t make any sense to me. Computer hardware has a 5 year depreciation schedule, so in fact you could take deductions every year for those servers you purchase, reducing your tax liability. How exactly are you paying taxes on that money?
You pay taxes on income. If you are saving money by not purchasing hardware the only thing happening is that you’re not able to reduce your taxable income through a depreciation deduction.
December 1st, 2006 at 4:57 pm
[...] Amazonのウェブサービスユニット（過去エントリはこちら：Mechanical Turk、Simple Storage Service[S3]、Elastic Compute Cloud [EC2]）は今年大きな波に乗っており、有力なスタートアップ企業が多数このサービスを利用して展開所要時間の短縮と節約に役立てている。TalkCrunch最新インタビューではAmazon CEOのJeff Bezosが同社のウェブサービス部門の戦略について語っている。 [...]
December 1st, 2006 at 6:15 pm
[...] Amazon Web Services (blog) make things like The Sheep Market, SmugMug, UnSpun, and NowNow possible. We’re happy to welcome them on board as a sponsor. Thanks! [...]
December 1st, 2006 at 11:19 pm
[...] Les services Web d’Amazon (dont Mechanical Turk, Simple Storage Service (S3) et Elastic Compute Cloud (EC2) que nous avons couvert) ont fait du bruit depuis leur lancement et pas mal de startups en vues les utilisent afin d’accélérer le déploiement d’applications et faire des économies. En France aussi d’autres startups utilisent ces services. Dans une interview récente Jeff Bezos discutait avec nous de la stratégie de sa société autour des services Web. [...]
December 2nd, 2006 at 11:52 am
Photagious and the state (or lack thereof) of web 2.0
Web 2.0 and PhotagiousWe are in the middle of what many are calling web 2.0. You will also find numerous postings on various blogs about bubble 2.0.As a disclaimer, I am a co-founder of FotoFlix where we have developed a…
December 4th, 2006 at 10:56 am
@Ted & Brad:
I know, sounds crazy, doesn’t it? But go ask a tax professional and they’ll tell you. The issues is you can only depreciate a certain amount each year, and you’re taxed on the rest.
Over 5 years, the net result is no taxation, but you pay a huge amount up front in year one and then get it back slowly over the next 4.
December 5th, 2006 at 8:59 am
[...] Amazon S3: Show me the money [...]
December 7th, 2006 at 7:19 am
[...] Amazon S3: Show me the money A good high level breakdown of the savings made by SmugMug from using S3 to store their data rather than building their own infrastructure. Worse yet, Amazon as still making money on this due to their economies of scale. (tags: amazon s3 work) [...]
December 8th, 2006 at 2:21 am
[...] Amazon’s web services unit (see our discussion of Mechanical Turk, Simple Storage Service (S3) and Elastic Compute Cloud (EC2)) has been making waves this year, and a number of high profile startups are using one or more services to speed deployment time and save money. In a recent interview, Amazon CEO Jeff Bezos discusses Amazon’s strategy around their Web Services unit. [...]
December 8th, 2006 at 3:27 am
I would be interested to know the storage to access ratio (i.e. how much network traffic do you generate for each TB of storage). Maybe it has already been mentioned somewhere, but I couldn’t find it… Can anyone help out?
December 15th, 2006 at 2:03 pm
Yea, it is great, s3. We are about to roll out a product that integrates with skype that allows for p2p and offline binary streaming. S3 will be a big part of the off line storage. We are very pleased with this. I am amazed that I can build a backbone platform off two existing networks, s3, and skype, to allow users to push and pull dynamic binary content. Lets get one think straight: as information becomes more dis-similar and divergent, services like s3 will continue to grow.
December 15th, 2006 at 2:07 pm
oh yeah, late congrats on the kids, we are expecting our fourth any day now…enjoy the blessings!
December 18th, 2006 at 4:20 am
[...] A recent entry on the SmugMug blog, Amazon S3: Show me the money, (via Steve Eichert) highlights the advantages of outsourcing one’s IT needs to Amazon’s S3 services. Businessweek covered Amazon’s web service strategy recently in Amazon’s Risky Bet. [...]
December 19th, 2006 at 3:29 pm
[...] SmugMug saves with S3 “Storage as a service” is more efficient in many ways than buying the disks. With all apologies to my friends in the storage business, this is the future for entrepreneurs. Some of the startup ideas I’m playing around with are beginning with the assumption that processing power and storage are essentially free. Ignore those constraints and much more becomes possible. [...]
January 5th, 2007 at 4:01 pm
[...] If you haven’t already you should really checkout the SmugMug cost saving discussion which included the following quote: But wait! It gets even better! Amazon has been so reliable over the last 7 months (considerably more reliable than our own internal storage, which I consider to be quite reliable), that just last week we made S3 an even more fundamental part of our storage architecture. [...]
January 8th, 2007 at 11:37 am
[...] Amazon’s web services unit (see our discussion of Mechanical Turk, Simple Storage Service (S3) and Elastic Compute Cloud (EC2)) has been making waves this year, and a number of high profile startups are using one or more services to speed deployment time and save money. In a recent interview, Amazon CEO Jeff Bezos discusses Amazon’s strategy around their Web Services unit. [...]
January 11th, 2007 at 4:11 pm
[...] So why should we doubt that the hardware equivalent of SaaS has similar potential? If the latency between “I need storage” or “I need a server” is a credit card and downloading and installing something like JungleDisk, well, how could that not be of interest? Even on a consumer level that has potential; I’m trying to persuade a researcher friend that rather than buy an external hard drive for some sizable GIS data that he has accumulated, he should just put it in S3. And for the start-ups discussed above? This could - and already is, in some cases - save them tens of thousands of dollars in hardware and hosting expenses. Will it pay off for Amazon? I don’t have even approximate figures for what it takes them to service an individual account, so I have no way of knowing, but I can say that I’m likely to go from paying them 17 cents a month in ‘06 to 17 bucks a month or more in ‘07 for business/personal use. [...]
January 12th, 2007 at 7:46 pm
Well, let me know if you want to offload some of those RAID arrays…
So, do you download back from S3 every image that your users look at (assuming you might cache some locally)?
January 16th, 2007 at 1:55 am
[...] You can serve just about any bits you want — see notes on using S3 to serve media files, serve BitTorrent, serve photos for a photo-sharing startup, and serve SecondLife client installers. [...]
January 16th, 2007 at 8:00 am
Hi,
I’m putting together an ad supported service that will generate images for users and then host them on S3. I’m somewhat concerned that a single user could send a large image out as spam, or generate an ad image and use up huge amounts of bandwidth (that I’ll have to pay for). Do you use any throttling or bandwidth caps on the images you host through S3? I googled for S3 customers and this entry was at the top of the list :).
Thanks!
Dan
January 20th, 2007 at 2:16 pm
[...] Original post by onethumb and software by Elliott Back [...]
January 21st, 2007 at 9:42 am
hello
January 22nd, 2007 at 3:39 am
[...] SmugMug has also become the poster child for Amazon’s S3 storage service, famously claiming to save at least $500,000 per year by switching. [...]
January 22nd, 2007 at 9:45 am
[...] SmugMug has also become the poster child for Amazon’s S3 storage service, famously claiming to save at least $500,000 per year by switching. [...]
January 22nd, 2007 at 11:45 am
[...] SmugMugのCEO、Dan MacAskillは過去に「Web 2.0なんて用語はどうも理解できない」とコメントを寄せていたが、彼のSmugMucgはWeb 2.0の専門家であるはずのわれわれにも、Web 2.0とは何かについていろいろと教えてくれる存在だ。同社は今夜（米国時間1/22）、いくつかの素晴らしい機能をローンチした。SmugMugが最初にユーザーを受け入れたのは2002年の末だったが、以来、本格的な写真家の愛好するサイトになっている。有料サービスを契約しているユーザーが10万人おり、1億枚の写真がアップロードされている。従業員は19人、十分に利益を上げており、過去一度も外部資金を調達したことがない。収入は年間$10M（1千万ドル）の大台に乗っているとMacAskillは私に語った。このサービスには無料版は存在しない。ユーザーはこのサイトにアップロードするためには最低40ドルの年間料金を支払う必要がある。プロ版の契約は年間150ドルで、作品にすかしを入れるなどのツールが提供され、写真家は作品のプリントアウトやダウンロードを販売することができる。上級のアカウントになると、ユーザーはテンプレートを利用して自分のアルバム（SmugMugでは「ギャラリー」と呼んでいるが）のルック＆フィールを完全にカスタマイズすることができる。またユーザー独自のドメイン名も利用できる。このサイトには典型的なWeb 2.0サービスであるFlickrにあるような、RSSフィード、タグ、コメント、オープンAPI、その他の機能を網羅している。Flickrの親会社であるYahooはSmugMugチームと定期的にミーティングを行っており、Yahooのユーザーインタフェース担当チームは最近YahooのブログのひとつでSmugMugを賞賛している。多くの(もちろん全部ではない) 真剣な写真家にとってSmugMugは自分の写真をプロモーションするための強力な手段となっている。SmugMugはまたAmazonのS3ストレージサービスの模範的ユーザーでもある。このサービスに乗り換えることで年間最低50万ドルが節約できたという話は有名。SmugMugの新機能今夜、SmugMugはいくつかの新機能を追加すると同時にシステムのアーキテクチャにも変更を加えた。実質的には完全なリニューアルに近い。SmugMugの95%のページが何らかの影響を受けている。バージョンアップの中心は、サイトの写真閲覧のインタフェースを（部分的にAjaxコンポネットを導入した）HTMLから、完全なダイナミックjavascriptに書き換えたことである。写真をクリックすると、従来のようなページの再読み込みは一切起こらず、ナビゲーションが大幅にスピードアップされた。また新しい写真をクリックするたびにページのトップに飛ばさることもなくなった。またYahoo Mapsと同様、SmugMugは訪問者がサイトを閲覧する際に、非常な手間をかけて、URLをアップデートする機能を追加した。これはリッチ・インターネット・アプリケーションに付き物の問題点で、FlashやAjaxアプリを利用してサイトを閲覧すると〔ページ再読み込みが行われないため〕URLが自動的に更新されない。SmugMugはこの問題を、Safariについてまで解決した。( Yahooでは依然として、Safariについては解決されていない)。SmugMugはまたデザインを変更して、ユーザーコメントを写真と同一のページに掲載することにした。（クリックが省ける）。さらに写真のメタデータや、表示オプションなどの大部分をマウスのフライオーバーによる表示とした。これによって多数のデータやリンクでページが散らからず、内容の写真が強調されることとなった。SmugMugの将来MacAskillによると、彼は常にベンチャーキャピタリストや買収希望者にアプローチされている。しかし彼は今後とも外部からの干渉なしに収益を上げる会社を運営していきたいとしている。彼は「私はプロの写真家のためにベストなサービスをこれからも提供し続ける。会社を売るつもりは全くない」と語った。もちろん、これによってベンチャーキャピタリストや買収希望者にいっそうSmugMugは魅力的に映ることになる。ところでSmugMugはWeb 2.0企業だろうか？ あらゆる点から答えはイェスだが、MacAskillがそれを認めることはあるまい。少なくともここ当分は。[原文へ] Smugmug [...]
January 25th, 2007 at 4:31 am
Don - thanks for sharing this and it’s great to see you doing so well.
Question (apologies if you have already answered it, I looked): does SmugMug use S3 for your ‘frontline’ image serving - ie for the thumbnails and photos I see on your webpages?
We are experimenting with using S3 for serving images, (for a completely unrelated application to SmugMug), and finding it just a little too slow to deliver the snappiness we seek. Your pages, even when viewing from Europe, are very fast.
Thanks
- Matt
January 26th, 2007 at 8:02 am
[...] SmugMug has also become the poster child for Amazon’s S3 storage service, famously claiming to save at least $500,000 per year by switching. [...]
January 26th, 2007 at 6:06 pm
[...] Vai alla pagina segnalata. [...]
January 30th, 2007 at 4:29 pm
[...] So what was the difference? We’ve been playing with using Amazon in a variety of different roles and scenarios at SmugMug. At first, we were just using them as a backup copy. That provided some great initial savings and a great deal of customer satisfaction as our customers became aware that their photos were safer than ever. As time went on and we grew more confident in Amazon’s ability to scale and keep their systems reliable, though, we moved Amazon into a more fundamental role at SmugMug and experimented with using them as primary storage. The week we started to experiment with that was the first of the two performance issues, and shined a bright glaring light on the downsides of using them in this way. We quickly shifted gears and are now quite happy with our current architecture, both from a cost view and a reliability view. [...]
February 2nd, 2007 at 12:43 am
[...] SmugMug est aussi devenu le “succès client” principal du service de stockage S3 d’Amazon, déclarant publiquement qu’ils économisent au moins 500.000$ depuis qu’ils l’ont adopté. [...]
February 23rd, 2007 at 3:09 am
good URL!!!!!!!!!!!!
February 24th, 2007 at 2:04 am
[...] Dan MacAskill (CEO of Smugmug) is their biggest protagonist. Businessweek had a cover story about it. Uzanto (Amit Ranjan’s company) uses it for SlideShare. Yes, I’m talking about Amazon’s Simple Storage Service (S3 for short). I think Amazon S3 is a brilliant idea and a boon for startups. It is not only cheap but, it also saves you all the effort of administration, backups, quality of service and the headaches of maintaining the infrastructure. The founders can focus on the business aspect of the idea and worry less about infrastructure. Here’s another great example about Amazon Web Services — this is the first example I’ve come across that uses several other services than just S3 — Blogarithms » Amazon for Infrastructure-on-Demand [...]
March 24th, 2007 at 2:02 pm
[...] Amazon S3: Show me the <b>money</b> [...]
April 5th, 2007 at 3:32 am
The site looks great ! Thanks for all your help ( past, present and future !)
April 10th, 2007 at 2:54 pm
[...] Yep - geek stuff. This is huge, in my humble opinion. For those of you new to the subject, S3 stands for “Simply Storage Service”. Amazon has a huge server farm and I think they realized that they have more disk space than necessary for Amazon. Some smart guy said “Why don’t we rent the available storage space to others?”. And thus, S3 was born. You can sign-up for an account to make use of the S3 services and use their API to create applications that store/retrieve “data” on/from their servers. Whether you’re a Python, Perl or Java guy, you’ll find modules that allow you to access the service transparently. This is all cool for geeks, but takes some time for the general audience to grasp. As such, the number of S3 applications is pretty small (SmugMug is one of the bigger names on the list of companies who use S3). [...]
April 17th, 2007 at 2:25 am
[...] Income from S3 is little more than a rounding error for Amazon, with nearly $11 billion in 2006 revenue. But the service has some passionate users who are claiming to be saving a lot of money versus handling storage themselves. It’s not too early to say, as Dan does, that “Infrastructure as a service has arrived.” [...]
April 17th, 2007 at 3:10 am
[...] Income from S3 is little more than a rounding error for Amazon, with nearly $11 billion in 2006 revenue. But the service has some passionate users who are claiming to be saving a lot of money versus handling storage themselves. It’s not too early to say, as Dan does, that “Infrastructure as a service has arrived.” [...]
April 17th, 2007 at 12:46 pm
[...] 今日は私はベンチャーキャピタリストと口喧嘩したり、TechCrunch20カンファレンスの宣伝をしたりするのに忙しくしている間に、Dan FarberがWeb 2.0 Expoのニュースをカバーしていた。ここでなかなか面白い事実が明らかにされた。AmazonのJeff Bezosがちょうど13ヶ月前にローンチしたオンデマンド・ストレージ・サービスのS3に保存されたアイテムが2006年7月の8億件から成長して50億件を超えたと発表した。S3サービスからの収入は Amazonの2006年の$11B（110億ドル）の収入にとっては四捨五入の誤差の範囲くらいのものだ。しかしこのサービスにはデータのストレージ処理を自前で行うのに比べて大幅にコストを節約できたとする熱心なユーザーがいる。Dan Farberが言うように、「インフラストラクチャー・アズ・サービス」の時代が来たといっても過言ではないだろう。ただし、Amazonとしては経費について注意する必要があるかもしれない。 昨年末のBusinessWeekの記事では数字を詳しく挙げてこの点を論評している。投資家にとっていちばん気がかりなのは、ここ３年以上続いたAmazonの新しいテクノロジーへの大盤振る舞いの投資である。現在、今年のテクノロジーへの投資は52%増加して$485M（4千850万ドル）に上っている。新しいテクノロジーを利用した新しいサービス開発のために数百人の技術者、プログラマーが採用され、それを稼動させるサーバー群が購入されている。その結果過去１年の営業利益率は4.1%とWal-Martの5.9%を下回っただけでなく、Webのおかげで旧態依然なものになったと考えられているbricks-and-mortar（オンライン店舗と対比した従来型の実店舗ビジネス）の書店チェーン、Barnes& Noble Inc. (BKS)さえ、もっと高い5.4%の営業利益率をあげている。「このハイテク投資はまだ全然利益に結びついていない」とPiper Jaffray & Co.のアナリストSafa Rashtchyは不満を漏らしている。 おそらくハイテクへの出費は投資家にとって最大の邪魔者だろう。われわれはAmazonの他のウェブサービスEC2とMechanical Turkを紹介している。また私は昨年11月、 Jeff Bezos とAmazonのウェブサービスグループについて話し合った。TalkCrunchでそのインタビューの模様が聞ける。[原文へ] Amazon s3 [...]
April 19th, 2007 at 3:30 pm
[...] Tym rewolucyjnym usługom, które miały premierę w zeszłym roku, od dawna przyglądam się z zainteresowaniem, ale aż dziwi mnie ich niezauważanie w naszym kraju. Zwłaszcza, że na świecie firmy z powodzeniem ich używają i chwalą sobie oszczędności dzięki nim poczynione. Jednym z największych pasjonatów i ewangelizatorów S3 jest SmugMug, który już pięć miesięcy temu twierdził, że dzięki S3 zaoszczędził pół miliona dolarów. To duży serwis, który pięć miesięcy temu miał ponad 110 tys. zdjęć, a biorąc pod uwagę ich tempo wzrostu w połowie tego roku powinni przekroczyć 200 tys. Wszystkie te zdjęcia są w bardzo wysokiej rozdzielczości, pozwalające na wydruk odbitek, w niektórych przypadkach całkiem sporego formatu. To wszystko brzmi jak dużo, bardzo dużo terabajtów danych. W 2007 roku Smugmug planuje zaoszczędzić dzięki S3 milion dolarów. [...]
April 22nd, 2007 at 10:24 am
[...] Income from S3 is little more than a rounding error for Amazon, with nearly $11 billion in 2006 revenue. But the service has some passionate users who are claiming to be saving a lot of money versus handling storage themselves. It’s not too early to say, as Dan does, that “Infrastructure as a service has arrived.” [...]
April 23rd, 2007 at 10:11 pm
[...] Although . . . I do see that SmugMug is using S3 just fine thankyou. Hmmm. [...]
April 25th, 2007 at 1:20 am
[...] Show me the money, Don MacAskill, Smugmug [...]
April 28th, 2007 at 5:06 pm
I completely believe this is the way of the world - with “server-less” hosting (or, in your case, disk-less storage). In turn, I’ve blogged a bit about the next revolutionary trend which involves Amazon’s EC2 and S3 at http://fountnhead.blogspot.com/2007/04/prediction-server-less-it-services.html
April 29th, 2007 at 11:07 am
[...] Amazon S3: Show me the money [...]
April 29th, 2007 at 9:35 pm
[...] Don MacAskill blogging about using S3 and how it’s saved SmugMug $500k. Damn. [...]
May 1st, 2007 at 1:54 am
[...] Here at CleVR, we use Amazon’s Simple Storage Service (S3) to host all of the panorama image files. This is a truly excellent service that helps us scale really easily without needing to buy loads more servers and RAID arrays and so forth. It’s great for small companies like ours, as well as not so small ones. The pricing is really simple: $0.20 per GB uploaded or downloaded, plus $0.15 per GB per month for storage. The guys at SmugMug reckon that they’ve saved $1 million in a year using S3. Anyway, one of their complaints was that there’s no volume discounts. [...]
May 1st, 2007 at 10:07 am
Amazon just released a new pricing model — I’m curious to see how this affects SmugMug!
May 1st, 2007 at 10:35 am
[...] May 1st, 2007 I had lunch yesterday with some of the fine folks at JanRain, and one of our discussions was about Amazon’s S3 … can a business actually save money, using it for file storage and distribution? It turns out there’s a few pretty good cases for it, the most impressive being SmugMug saving half a million bucks vs. their DIY approach. [...]
May 1st, 2007 at 12:29 pm
[...] Since we were already saving a ton of money using S3, this is music to my ears. tags: amazon s3, smugmug, storage, amazon [...]
May 1st, 2007 at 2:00 pm
[...] This is great news, I just got an email from Amazon Web Services with their new pricing. I wonder how much SmugMug will save now? Mr Zawodny’s going to have to redo his calculations now too. [...]
May 2nd, 2007 at 5:59 am
[...] SmugBlog: Don MacAskill » Blog Archive » Amazon S3: Show me the money — I personally used S3 to backup my entire iTunes library. I’ve got about 40GB of stuff on there and it costs me about $5-10 a month. Super! [...]
June 8th, 2007 at 10:10 pm
[...] Is SAAS old hat to you, already reaping the benefits? Looking for the next opportunity to concentrate on your core business and shed overhead? Well the next step may just be infrastructure as a service. Amazon has built an interesting offering around this, their two core products Elastic Computing Cloud (EC2) and Simple Storage Service (S3) allow for on demand consumption of core computing resources. Both of these services are in their early stages, where SAAS was a couple years ago, but the adoption among tech savvy small companies has been phenomenal (5 billion objects stored in S3). Some are even running their entire business off of it, smugmug.com. [...]
July 2nd, 2007 at 4:30 pm
[...] Michael Arrington de TechCrunch.com quedo sorprendido al borde de no creerlo cuando el CEO de Smugmug.com, un repositorio de fotografías similar a Flickr cuando le comento que el año recién pasado se habia ahorrado aproximadamente 500 mil dolares (256 millones de pesos chilenos) en el concepto de alojamiento y ancho de banda de las casi 110 millones de fotografías que hostean. [...]
July 8th, 2007 at 11:33 pm
What about the authentication problems with S3? Does smugmug have a special arrangement with Amazon? This has been addressed a couple of times, and all someone has to do is control your Email, reset your password, then wreak havok on all your S3 storage with the reset password. Can this really be?
One problems with S3 is that each operation requires the secret keys that are assigned by Amazon, and your users need to include these at the client. So, are you going to send the secret keys to every client? Anyone with a debugger could lift the keys in about one minute. The other option is to have the users upload or download to your own server, but the trouble with that is you incur BW cost twice, and worse. That’s where EC2 comes in. You can use EC2 to interface with each client, then the keys all stay put on Amazon. Transfers between EC2 and S3 don’t cost.
But, what about the Email authentication problem? Is the whole of smugmug really vulnerable to someone gaining control of a single Email account and attacking their S3 storage?
July 13th, 2007 at 12:37 am
[...] It’s also racking up a number of passionate users who swear by it for reliability and cost savings. Phanfare is just the most recent example, albeit a large one. [...]
July 13th, 2007 at 2:20 am
[...] It’s also racking up a number of passionate users who swear by it for reliability and cost savings. Phanfare is just the most recent example, albeit a large one. [...]
July 13th, 2007 at 4:13 am
[...] It’s also racking up a number of passionate users who swear by it for reliability and cost savings. Phanfare is just the most recent example, albeit a large one. [...]
July 13th, 2007 at 3:34 pm
[...] スタートアップのPhanfareは大量のユーザー生成メディアを保管しているが、今日（米国時間7/13）、40テラバイトに上る全バックアップデータをAmazon S3ストレージ・サービス に移すプロセスを開始したと発表した。Amazon S3は最近勢いに乗っているようだ。50億以上のオブジェクトを保管、さらに急成長を続けている。また信頼性の高さやコスト削減に非常に役立ったことを証言する熱烈な ファンの数にもこと欠かない。Phanfareもこういったファンの列に加わった最新のユーザーだ。もっとも非常に大きなユーザーではある。ただしPhanfareはデータをすべてS3に移すことは断念した。当面移動するのはバックアップだけだ。ストレージ機能のすべてをAmazonに移せば、さらに経費の節約になることは彼らも認めているのだが―結局、現在の段階では、AmazonはSLA（ Service Level Agreement 品質保証契約）を提供していない。それどころかサービスに不満があった場合にどこへ連絡すればいいのか、電話番号さえ公表していない。Amazonが顧客のデータを失くすようなことがあるとは思えないが、われわれとしては顧客から預かっているデータを保管させる以上、まずSLAが必要だと考える。Amazonについて言えば、現在のストレージ・サービスに加えて、年内に、ウェブ上でコンピューティング・サービスの提供を始めるという噂が流れている。さらにストレージとコンピューティングの両サービスを補完するMySQLを利用したデータベース・サービス も準備されているという。[原文へ] Amazon Amazon S3 Phanfare [...]
July 19th, 2007 at 5:12 am
[...] I have been amazed at how cheap reliable storage is these days with services like Amazon S3. The S3 introduction documentation doesn’t mention using python and django so I thought I’d implement the demo in django. It really is very simple to take advantage of S3 in your django apps, as Adrian shows here. [...]
September 14th, 2007 at 7:41 pm
[...] What surprised me most was the level of maturity and rapid/breadth-of adoption of AWS. First, the AWS ‘tires’ have already been ‘kicked’: the services are mature. The Internet Archive, for one, testified that the EC2 machines crash far less often and that the bandwidth has become faster/more-reliable. Second — and relatedly — there are already many more start-ups using AWS than I expected. Smugmug is the poster-child, and given that he started almost a year ago, I suppose I shouldn’t be that surprised that AWS is catching on so rapidly … but, just, wow, hearing the VCs say that they consider it with every investment and are actively looking for companies that are using AWS … this stuff is mainstream! [...]
September 14th, 2007 at 8:14 pm
[...] Don MacAskill, CEO of SmugMug, who I’ve corresponded with a bit through this site, says SmugMug has saved $1 million dollars by using S3 (and continues saving). [...]
October 17th, 2007 at 11:08 am
[...] Marc Liron - Microsoft MVP http://www.marctalkstech.com Source: http://blogs.smugmug.com/don/2006/11/10/amazon-s3-show-me-the-money/ [...]
October 17th, 2007 at 11:45 am
SmugMug.com Use Amazon S3 To Save $500K
Yep that’s right!
In 2006 SmugMug.com used Amazon S3 Web Service to save $48,490 / month… ($500K) on their business costs.
…and they expect their savings from Amazon S3 to be well over $1M in 2007, maybe as high as $2M.
So…
January 23rd, 2008 at 6:23 pm
[...] are some who would obviously dispute my conclusions, such as SmugBlog (marketing ploy?) and actual small blogs (not to mention my partner in crime, Adam) These icons [...]
January 29th, 2008 at 4:06 pm
[...] you choose to go with a storage service (SaaS) rather than an internal system. Most notably, SmugMug uses Amazon S3 for this. This is a viable option, and while I am a fan of Amazon S3, users need to read [...]
February 4th, 2008 at 12:46 am
Smugmug is one of the companies I admire most. Your product offerings, customer service, innovation and focus on your core product quality is … well, super-excellent and awesome. I’ve hosted http://photos.rajiv.com/ on Smugmug for several years and continue to be very pleased with Smugmug.
For my primary web site, http://www.rajiv.com, I’ve moved from a regular web host to Amazon EC2 and S3. Reading this blog post was one of the sources of inspiration.
Thank you!
– Rajiv Pant
New York, NY
February 11th, 2008 at 9:11 pm
[...] Amazon S3: Show me the money [...]
February 18th, 2008 at 6:02 am
[...] I was running down the cause of Friday’s S3 service interruption, I came across this excellent financial analysis of the cost savings of a web services deployment versus that of a traditional platform approach. [...]
April 4th, 2008 at 10:44 am
Sorry, but this is a bit misleading. You weren’t doing things well to begin with, plopping in server after server with disk arrays attached locally. You can’t compare the cost of doing things the wrong way to the cost of S3 and call that an accurate comparison. I work with those “over-priced storage solutions” and I can implement a highly-scalable cost-effective solution cheaper than Amazon prices, including the equivalent monthly charge for bandwidth, power, etc.. You could probably do it even cheaper still looking at something like Isilon - which Kodak uses for their EasyShare service.
Amazon S3 can make sense for small or mid-size companies, I don’t disagree with that. In many cases, the people resources don’t exist within the company to support running even a small data center. But your argument was based solely on hardware implementation costs, and your comparison is not a good one.
At the end of the day, one rule still applies: You get what you pay for. Not only can I implement a solution cheaper than Amazon with those “over-priced storage vendors”, I can implement one that doesn’t go down 4 times in 2 years - http://blogs.smugmug.com/don/2007/01/30/amazon-s3-outages-slowdowns-and-problems/
April 4th, 2008 at 10:51 am
@Not Apples to Apples:
Prove it. I have a lot of money to invest in storage, and if you can really deliver a solid storage solution with $0.15/GB/month TCO, including a minimum of 3 geographically replicated copies (so $0.15/3GB/mo), and which will continue to drop in price without leaving me stuck with sunk costs, I’m all ears and the contract is yours. Better yet, prove that you could have done this 18 months ago (I expect an Amazon price drop “soonish” since disks continue to get cheaper).
I’ve looked at nearly every provider there is, including Isilon, and they’re all much much more expensive once you factor in replicated geographically disperse copies, including bandwidth, power, and man hours.
But I’d be thrilled to be proven wrong - and you’d win a multi-million dollar contract. Make sure your pitch is truly apples to apples, of course.
So put up or shut up
April 5th, 2008 at 10:19 am
[...] and final thing I’ll get into. I won’t get into any numbers…you can read about some examples on Smug Mug’s blog and Digital Web Magazine’s article about Blue Origin. However, what I can say is that yes - you can [...]
April 8th, 2008 at 9:30 pm
I think we are all forgetting one very important fact. S3 is not a file system. S3 is storing binary blobs attached to a URL. This might work well for some applications, but not for all. So the comparison is not exactly apples to apples by nature.
This gives Amazon the opportunity to spread around the disk across many CPU’s (”Google File System” style) most likely utilizing commodity hardware (read much cheaper as far as hardware goes). Now in order to do this economically it requires a lot of man power (maintenance), which is possible only on large scale.
I personally do believe “@Not Apples to Apples”. You can get a price of storage alone < $0.15/GB on a fairly small scale. Not sure you fit in that category. Can you site the total amount of storage you have?
Now I doubt this can be done using hardware from the big boys (EMC, etc). I would love to see numbers that say otherwise.
As you stated your pre-existing setup was close to $6/GB, but that included RAID. Discounting for RAID 5 (losing 20%) this means about $4.80/GB. The price of Isilon is around $1.5/GB, the price of Coraid is about $0.71/GB (they claim $0.64) - http://www.coraid.com
So this is 3 to 6 times lower.
BTW: The Isilon solution is very scalable and nice, but the main advantage is you get a big NAS device within one file system, which I don’t think your application cares about.
All the prices above however are purchase price for hardware. To compare apples to apples with S3 we have to provide RAID and double the number in order to provide a second copy. Then we have to lease the equipment and account for the cost of financing. This still does not account for the price of replication (at a minimum the cost for bandwidth). Let’s do the calculations for a typical 36-month lease:
If the purchase cost of storage hardware per RAW GB is $1, the true cost will be $1 / 0.8 * 1.10^3 = $1.66375 for 3 years = $0.0462 per month
This assumes RAID 5 (N+1) with 5 disks per array (the 0.8) and 10% per year interest on the financing
So the magic constant for converting purchase cost to monthly cost under these conditions is 0.0462
Now we need to multiply by 2 for redundancy. So for $0.71/GB purchase price we get about $0.0657/GB
You are left with 8.5 cents per GB to pay for everything else, which should be doable. The everything else here includes rack space, power, warranty and repair cost and labor for support.
None of this accounts for other infrastructure. 1 Gbps switches are quite expensive. Not to talk about 10 Gbps.
The way I came up with $0.71/GB for the Coraid box was to spend $4000 for SR1521 Appliance + $4000 for 16 x Seagate Barracuda ES.2 ST3750330NS 750GB at $250 a piece. This leaves you with 1 spare HDD. Total is $8,000 for 11.25 TB.
Notice the cost of hard drives equals the cost of CPU. Now consider if you used commodity hardware. You can assemble a similar configuration for about $2,000 and now the CPU to storage cost becomes 1:2
This obviously is not the same as the Coraid box. It requires additional labor to assemble, test, create OS image, etc.
You could even get BUFFALO 4TB TeraStation Pro II. Not very scalable, but purchase price is $0.50/GB. Granted the RAID loss here would be 25% (only 4 drives).
Now coming back to the initial issue. You are not in the business of managing storage (and other infrastructure), so IMO you are better off with S3. The only thing that has to be considered carefully here is bandwidth cost. $0.20/GB is very expensive. If you start using S3 as the space where you provide data to end users they could sink you. End users in your case are individuals who typically have unlimited download, so they won’t care if they download the same file 15 times. It would be different than using it for backup.
You probably already have good statistics on the amount of bandwidth consumed by end users, so take a careful look there.
April 9th, 2008 at 12:00 am
[...] Ejemplos: Smugmug Servicio para compartir fotografías en línea las cuales las almacena en los servidores de Amazon, asegurando que con ello lograron un ahorro de 1 millón de dólares. [...]
April 19th, 2008 at 7:51 am
[...] go away. In fact it did just go away because one day I discovered Amazon S3, moreover I read on Don MacAskill’s blog that his company switched over in about a day. So did [...]
April 21st, 2008 at 5:57 pm
[...] is very well documented. Smugmug a photo hosting service have posted amazon s3 savings in their blog. Amazon S3 can be accessed by web browser from this [...]
April 22nd, 2008 at 10:18 am
[...] SmugMug [...]
June 4th, 2008 at 1:46 pm
Looks like Nirvanix has package prices on unlimited transcoding services.