Friday, January 20, 2012

Amazon DynamoDB: First Impressions

An e-mail arrived this morning from Amazon Web Services (Aws):

It was a new feature announcement from Aws.

The new product is DynamoDB... a new NoSQL database engine for Amazon Web Services (Aws)

From the e-mail, it's described as "Amazon DynamoDB, a fully managed NoSQL database service that provides fast and predictable performance with seamless scalability".

It has some interesting key points:
  • Data on Solid State Drives (SSDs)
  • Replicated data synchronously across multiple zones.
  • Supports two types of primary keys: Hash Type and Hash and Range Type 
  • JSON for data transport
This new product is a NoSQL database engine built into their cloud infrastructure (similar to SimpleDB as a product, but with some notable changes and upgrades).

From CTO Werner Vogels' DynamoDB blog post he states the differences between SimpleDB and DynamoDB: "While SimpleDB has been successful and powers the applications of many customers, it has some limitations that customers have consistently asked us to address." Listing: Domain scaling limitations, Predictability of Performance, and SimpleDB's Pricing complexity.

From the documentation, it appears to be similar to their existing  SimpleDB offering, but is designed for large scale needs (and predictable) performance. Most notably no limits or restrictions on database size, and automatic management of scalability across multiple servers.

What is NoSQL? 
According to Wikipedia's  NoSQL article:
In computingNoSQL (sometimes expanded to "not only SQL") is a broad class of database management systems that differ from the classic model of the relational database management system (RDBMS) in some significant ways, most important being they do not use SQL as their query language.
DynamoDB Storage Model (1000 ft view)
From their documentation, each object (or item) is stored within a table, with the individual values within the item stored as attributes:
From Amazon DynamoDB (beta)
Getting Started
In order to get started with DynamoDB, you have to make sure you have a couple of things:
  • Amazon Web Services developer account
  • Your AWS account has to be set up to use Amazon DynamoDB
  • Your AWS Access and Secret Key (in order to run the examples, or start your development project)
DynamoDB and .NET
The AWS SDK for .NET has been updated to contain samples for using DynamoDB with the .NET framework (fair warning if you have a previous version of the SDK you may have to remove it to install this one).

The SDK contains (basically) 4 samples:

  • Data Model Sample - creation of a context, persistence of items using the context, and some queries against the context.
  • Data Model Schema - Samples of objects that can be stored within DynamoDB. Specific would be the use attributes: DynamoDBTable that "Specifies that this object can be stored in DynamoDB". DynamoDBHashKey, DynamoDBProperty, and DynamoDBRangeKey to name a few.
  • Document Model Sample - Demonstrates the document model. Shows creation of a table, and adding documents to a table, persistence, and querying from the model.
  • Table Operations - Basic table operations: Contains, CreateTable, table status, table deletes.

These samples are available for VS 2008 and VS 2010. With my installation, the samples were located at C:\Program Files (x86)\AWS SDK for .NET\Samples\AmazonDynamoDB_Sample.

Pricing (as of 1/19/2012)
Notes About Amazon DynamoDB states it pretty well "pricing is based on actual write/read operations and not API calls (e.g. a query returning 100 results accounts for 100 ops and not 1 op)".

You pay a flat, hourly rate based on the capacity you reserve:
Throughput Capacity
Write Throughput: $0.01 per hour for every 10 units of Write Capacity
Read Throughput: $0.01 per hour for every 50 units of Read Capacity

The listed pricing (as of 1/19/2012) offers a free-pricing tier as well:
Free Tier*
As part of AWS’s Free Usage Tier, AWS customers can get started with Amazon DynamoDB for free. DynamoDB customers get 100 MB of free storage, as well 5 writes/second and 10 reads/second of ongoing throughput capacity.
Final Thoughts:
... What's missing: From CTO Werner Vogels' DynamoDB blog post one comment from Faraz points out one of the key components missing from DynamoDB's initial offering: "missing critical database pillar of snapshot backup and fast recovery system". Vogel's response: "we have a philosophy of launching with a minimal feature set and then quickly iterating while prioritizing based on customer feedback. Backup/Restore for DynamoDB will have high priority". While true, I'm still waiting for micro in VPC!

... "pure local emulation is not available".  However, there is an open-source project to provide this emulation for SimpleDB: fakesdb (a fake version of Amazon's SimpleDB for local/integration testing).

... JSON is used for sending data and for responses, but it is not used as the native storage schema (from
Notes About Amazon DynamoDB).

... DynamoDB vs. Cassandra

No comments: