Posts Tagged: amazon

General — View Comments
27
Oct 09

Amazon launches MySQL in the cloud

 logo_awsAmazon has launched Amazon RDS (Relational Database Service), as part of its AWS offering.

Amazon Relational Database Service (Amazon RDS) is a web service that makes it easy to set up, operate, and scale a relational database in the cloud. It provides cost-efficient and resizable capacity while managing time-consuming database administration tasks, freeing you up to focus on your applications and business.

Amazon RDS gives you access to the full capabilities of a familiar MySQL database. This means the code, applications, and tools you already use today with your existing MySQL databases work seamlessly with Amazon RDS. Amazon RDS automatically patches the database software and backs up your database, storing the backups for a user-defined retention period. You also benefit from the flexibility of being able to scale the compute resources or storage capacity associated with your relational database instance via a single API call. As with all Amazon Web Services, there are no up-front investments required, and you pay only for the resources you use.

Similar to its EC2 service, you have to launch an RDS instance, which costs between $0.11-$3.10 an hour (depending on the number of dedicated cores and memory you need — from 1 core with 1.7GB of memory up to 8 cores with 68GB of memory).

It would have been nicer if Amazon would have launched this as pay-for-what-you-use service, similar to its other SimpleDB and S3 offerings.

General — View Comments
3
Sep 09

Jumping Clouds

rackspacecloudI have been a Dreamhost customer for a number of years now, but because I’d like a bit more control over the versions of software (PHP, MySQL, memcache etc.) I use I decided to move to a VPS again. There are many good ones out there, but I decided to go with the Rackspace Cloud. Attempting to compete with Amazon AWS, Rackspace is rolling out a competing cloud platform. Currently it consists of Cloud Servers, Cloud Sites and Cloud Files, but given the fact that they hired the developer of Cassandra, a Cloud Database might be forthcoming.

Cloud Servers is what I’m using now, it basically allows you to cheaply spin up servers (virtual private servers, to be precise). You get a choice of a number of Linux distributions that you have to manage yourself. Similar to Amazon, you pay per hour for these servers. One big difference is that the lowest end Amazon EC2 server costs roughly $70/month and includes 1.7 GB of memory and 160 GB of hard drive space. Cloud Server configurations start at 256 MB of RAM (with 10 GB of hard drive space) which cost roughly $10/month, and configurations upwards to 15 GB of RAM (with 620GB of hard drive space) that costs $700/month. The nice thing is that you can easily up and downgrade your servers without the needs to reinstall. You can make back-ups which (soon) will be stored on Cloud Files (see below), when you instantiate a new machine, you can start it up from one of your back-upped images. Of course a $10/month VPS is very competitive with regular hosting, especially considering the fact that you get root access and thus can install whatever you want. And, because you only pay per hour, you can even just try it for a little while to see if you like it.

Cloud Sites is a more managed solution than Cloud Servers which starts at $100/month. It is mainly aimed at high-traffic sites, such as gdgt.com (which is hosted by them). Sites hosted on Cloud Sites are in fact hosted on a large number of machines, a number that is dynamically expanded as traffic increases. The nice thing about this solution is that you don’t have to worry about scaling your website yourself. The disadvantage is that the cost is higher and you do not get root access to the servers.

Cloud Files is a direct competitor of Amazon S3 (Simple Storage Service). You upload your files there, store them for $0.15/month and they are distributed over Limelight’s content-delivery network. Outgoing bandwidth through this CDN costs $0.22/GB.

Main — View Comments
23
Jan 09

The hashtable is the database

hashtableIf you pretend to be at least a little bit hip, you dumped your SQL database server and moved back to the core. To the essence. The format that fits any problem. No, it’s not the string of bits, lets not go that far (yet). It’s the hashtable. The hashtable is the anti-relational database. You might think it’s highly impractical or limited, but that’s where you’re wrong. What matters is not practicality, what matters is scalability. It turns out that hashtables scale like butter. A few days ago I talked about project Voldemort, used by LinkedIn. Similarly, Amazon has Dynamo. CouchDB is not that much more than a hashtable either.

“But what if I want to do X, Y and Z? I can’t do that with a hashtable!”

If you can’t do X, Y and Z with a hashtable, there must be a good reason for it. As you develop X, Y and Z you are likely to find they were bad ideas anyway. If features don’t fit the hashtable model, more often than not they turn out to be pretty crappy features.

Main — View Comments
14
Jan 09

The project that shall not be named

voldemortProject Voldemort is an open source implementation of parts of Amazon’s Dynamo (on top of which Amazon built various of its AWS web services, such as S3). It’s a database in its most basic form: (key, value) pairs. Which reminds me of a visitor we once had in our group. He visited because he developed software that generated code (which is what we do research on). It turned out that he used PHP to generate HTML code, which is hardly ground breaking. Nevertheless, he had an interesting view on database design: “all this normalization and column type stuff in database seems pointless to me,” he said. “In the end, all you need is strings. That’s why all columns in my tables are of type string with a fixed column width.” Looking at current developements, this statement could almost be seen as prophetic. The database 2.0 that starts to emerge indeed seems to move back to denormalized tables and String column types. Look at Amazon SimpleDB and CouchDB and you’ll see what I mean.

But I digress. The key, value pairs in Voldemort are automatically distributed over multiple server to make the system scale. Voldemort is used by a little site called LinkedIn that you may know. Why I care? Because Harry Potter is cool and Voldemort is a kick-ass name for a software project.