Slashdot Mirror
Amazon Beefs Up Its Cloud Ahead of MS Announcement
Amazon has announced several major improvements to its EC2 service for cloud computing. The service is now in production (no longer beta); it offers a service-level agreement; and Windows and SQL Server are available in beta form. ZDNet points out that all this news is intended to take some wind out of Microsoft's sails as MS is expected to introduce its own cloud services next week at its Professional Developers Conference.
Replace "cloud" with "mainframe" and take 40 years off your age, and then you pretty much have it, as is my understanding.
"The service is now in production (no longer beta)"
Then they have already reached a state that Google will never achieve.
Better known as 318230.
To expand on this, because now you've made me research this, basically cloud computing refers to hosting business applications remotely- typically, but not neccessarily, on multiple servers. (Such as an application server, a sql server, and so on)
"But I already have my business software hosted on an application server, and it utilizes a seperate SQL server... how is this any different?"
Is it stored somewhere offsite, say, by a hosting company?
"Why, yes.."
Then welcome to the cloud computing club!
But I've been doing this since the late 90s, I'm confused, what's changed?
Nothing at all. It's just like podcasts and web 2.0, another useless name for downloading audio files and websites that are more clever than before.
So basically, the only difference between remote hosting and cloud computing is whether or not you understand what's underneath the hood. If you're not sure how it works, but it just does, it's called "the cloud" otherwise, the rest of us call it "Shared hosting," "VPS," "Colocated," or "Dedicated" offsite hosting.
It's kinda like using the word magic instead of the word science. Makes people feel better.
Belief? Hope? Preference?The Existential Vortex
After reading the SLA at http://aws.amazon.com/ec2-sla/, I see it as all a big show with no real guts behind it:
# Availability is averaged over the last 365 days, but you only get credit for the current month's costs.
# You only get a service credit for 10% of the current month's costs. If you decide to move your business elsewhere, you may not apply the credit toward any past charges, including for the month in which the outage occurred.
# Availability refers to the "region" availability, and makes no guarantees about instance (computer) reliability, storage consistency/reliability. As far as I can imagine, it might be rather hard to figure out what constitutes a region's "availability" independently. The official measure stated in the SLA is basically a measurement made solely by Amazon.
# To receive any of this pathetic service credit (again, it is not a refund), you are required to send Amazon an email documenting (dates, times, regions) and providing evidence (heartbeat request logs, etc). *Yes, they want logs.* For almost all of their customers, the time and effort involved in filing a claim would outweigh the benefit of the credit.
It's not just the remote hosting that's appealing - it's the scalability.
If I write an app and put it on a dedicated host, I'm okay until I exceed the capacity of that host. Then I have to find another box or boxes and I may even have to change my software since I had assumed it would only be on one server. Finding additional capacity, refactoring and load balancing not only add cost, but effort (and therefore time).
On a service like EC2 (or even Google Apps), I'm renting space on the massive infrastructure of Amazon or Google. Their frameworks restrict you from developing anything that can only run on a single server. And if I need more capacity, I just right a bigger check that month.
That scalability goes for bandwidth as well. If you poke around the internet, you'll find lots of folks using Amazon's storage service for that reason.