Slashdot Mirror
Inside Amazon's Cloud Computing Infrastructure
1sockchuck writes: As Sunday's outage demonstrates, the Amazon Web Services cloud is critical to many of its more than 1 million customers. Data Center Frontier looks at Amazon's cloud infrastructure, and how it builds its data centers. The company's global network includes at least 30 data centers, each typically housing 50,000 to 80,000 servers. "We really like to keep the size to less than 100,000 servers per data center," said Amazon CTO Werner Vogels. Like Google and Facebook, Amazon also builds its own custom server, storage and networking hardware, working with Intel to produce processors that can run at higher clockrates than off-the-shelf gear.
working with Intel to produce processors that can run at higher clockrates than off-the-shelf gear.
What does this mean? They have custom chips? Custom mods at the chip fab level? Or are they taking advantage of designed-in features that are locked out for normal chip users? Are they simply over-clocking? Or are there features that can be unlocked with money?
If you want news from today, you have to come back tomorrow.
I've been unable to get IPv6, and Amazon's been promising "we're working on it" for years now....
The cloud went dark with only the blinky status light for company.
“Every day, Amazon enough new server capacity to support all of Amazon’s global infrastructure when it was a $7 billion annual revenue enterprise,” said James Hamilton, Distinguished Engineer at Amazon, who described the AWS infrastructure at the Re:Invent conference last fall. “There’s a lot of scale. That volume allows us to reinvest deeply into the platform and keep innovating.”
Did they use AWS for translation on this paragraph? How do you have "a lot of scale"? One can scale up or down, but is this like a computer hokey pokey? Scale is a verb!
Really, I skimmed this one pretty lightly. It looks like a marketing article, not a technical article. Buzz words a plenty, so I'm guessing your question is answered by "marketing"..
-The wise argue that there are few absolutes, the fool argues that there are no probabilities.
They are expensive and you have to buy a lot, but they'll do custom. Oracle also buys custom Intel chips. There are limits to what they'll customize, obviously writing a whole new ISA wouldn't be possible (at least not without a shit ton of resources) but they can customize things like cache sizes and configurations.
In terms of clock rate I image what Amazon is doing is more or less having Intel raise the TDP for the chips and run them harder. All the Xeons cap out at about the same TDP for the high end, regardless of core count, so higher core count chips are slower. However with aggressive cooling, you can have a setup that'll cool more than that TDP. So Amazon might contract to Intel to sell them higher rated chips, with the understanding of the increased cooling needs.
It's the fact that they only focus on infrastructure. IaaS is their bread and butter and it's what keeps them running and going with companies that don't know anything better than servers and storage, to migrate their workloads (the peaks and valleys kind) into the cloud to save money and be agile.
The next generation is a step beyond that, and it's what Microsoft, SalesForce and Google are building for -- PaaS. The idea that you manage fleets of servers is an archaic one, and the next generation will be writing against an API that manages all of that for you. Azure's Service Fabric, Google's AppEngine, SalesForce's Heroku -- those are the future of cloud computing. It's also a future that AWS doesn't have represented at all.
I am a fan of AWS technologies in their current state and the problem they solve for. But it's a problem that takes EXISTING methodologies and infrastructure and merely replaces them. It does not help prepare for the next generation of developers who grow up with the idea that this is all a commodity and they just want their code to work and execute, and have a smart engine behind it figure out all the needs for their app (be it data, network, power, cooling, memory, etc).
In that sense, Microsoft is far, far ahead of the others and as developers start to change their tune in their practices, we'll see that uptick for Azure happen. In the meanwhile, AWS is a decent place to put your existing servers and storage type of needs.
The price is always right if someone else is paying.
I'd say that the cloud has been very good for Intel. Wouldn't you?
Not to be too pedantic, but Crikey! That was very close to the worst edited article I've ever read - even on the web, which is saying a hell of a lot! C'mon, guys, you're supposed to be some kind of publication, for Christ's sake!
This looks impossible. How can the MS cloud division pay the OS division for the scale of computers used by Amazon? 80,000 servers x 30 = 2.4 million with data center licensing that would be about $10 billion! LOL That assumes they are 2 socket, if they are 4 socket servers Amazon/MS would have to pay $20 billion!
Your Average Joe
"As Sunday's outage demonstrates, the Amazon Web Services cloud is critical to many of its more than 1 million customers"
I thought the outage demonstrated the relative unreliability of Amazon cloud Services. What are the legally binding terms of services that AWS provide in relation to uptime.
“Every day, Amazon enough new server capacity ..."? Editors at datacenterfrontier.com please!!!
Tiller's Rule: Never use a word in written form that you've only heard and never read. You will end up looking foolish.
You didn't build out a Multi AZ solution for your critical app? You relied on AWS services for critical load balancing and fail-over? You shoved everything into US-EAST-1 where it can sometimes take 5 minutes for a reboot? You're doing it wrong.
Harrison's Postulate - "For every action there is an equal and opposite criticism"