The latest string of updates from AWS continues the cloud provider's attempts to streamline its services for customers...
that don't want to build everything themselves.
Amazon this week introduced a service called AWS CodeStar, which abstracts some of the steps needed to build a tool chain for continuous integration/continuous deployment. It also added capabilities to Amazon DynamoDB and Redshift that will reduce the time and money users spend on those services.
Redshift Spectrum can run SQL queries against exabytes of unstructured data in Amazon Simple Storage Service (S3). Redshift queries have been limited to structured data, but this service targets the growing number of customers who have moved vast amounts of unstructured data to S3 because of the low cost, Amazon said.
A connection between the services removes one of the biggest bottlenecks associated with big data projects.
"If you can use a data lake you're not going to have to load and unload the data based on your capacity," said Adam Book, principal cloud engineer at Relus Technologies, an AWS consulting partner in Peachtree Corners, Ga. "It's going to be a much more cost-efficient setup if you don't have to spin up new clusters and pay for those nodes."
Redshift Spectrum is currently limited to U.S. East (Northern Virginia), U.S. East (Ohio) and U.S. West (Oregon), though AWS said the tool will expand to other regions in the coming months.
Another attempt to improve a popular service is Amazon DynamoDB Accelerator (DAX), a fully managed in-memory cache that reduces response times from milliseconds to microseconds, even with millions of requests per second. Developers can add DAX to DynamoDB applications through the AWS Management Console and doesn't require application rewrites.
DynamoDB is typically used for web, gaming and mobile applications, but prior to DAX, IT pros had to deploy and manage in-memory caching clusters in front of DynamoDB applications that required near-real-time data access.
Every NoSQL database has its strengths and weaknesses, and one of the weaknesses of DynamoDB was how it worked with applications that have massive amounts of read-only workloads. DynamoDB could address that problem and provide big performance improvements for users, said Mike Kavis, vice president and principal cloud architect at Cloud Technology Partners, an AWS consulting partner in Boston.
Cache libraries aren't a major challenge to build, but it is repetitive to add an application layer to handle the database, added Kevin Felichko, CTO at PropertyRoom.com, an online auction company in Frederick, Md.
"It's nice [AWS is] seeing that design pattern and that they're making it easier for a DevOps setup where now this pattern exists and you don't have to keep recreating the wheel every time," Felichko said.
The one major drawback to DynamoDB Accelerator is its limitation to Java libraries, he said. He would like to see Node.js support, though he expects that to come as the service matures.
AWS CodeStar offers a quicker path to DevOps
CodeStar appears to be AWS' attempt to abstract at a much higher level so people don't have to worry about infrastructure and middleware and can instead focus on code, Kavis said.
"This sure looks like PaaS to me," Kavis said. "It looks like Heroku eight years ago."
CodeStar probably isn't the best fit for complex or custom applications, but it does follow a pattern from AWS. The company started with a sandbox approach that required a lot up front work by its customers, but over time, Amazon has added tools to reduce the burden for less IT-savvy development shops. Those services range from serverless frameworks, such as Lambda functions, to fully managed virtual private servers with Amazon Lightsail.
"[AWS] is giving people options," Felichko said. "If you want to go in and do it yourself you're free to do it, but if you ... don't want to go through all that stuff and time is urgent then you can add these pieces instead."
All of these updates were disclosed at the AWS Summit in San Francisco. The cloud provider also advanced the availability of several previously announced services:
- Amazon Lex, the AI software behind Amazon Alexa, was made generally available. The fully managed service can be used to build applications that incorporate speech recognition and natural language understanding.
- The EC2 F1 instances were made generally available in the U.S. East (Northern Virginia) Region, though it's unclear when the rollout will extend to other areas. These come with Field Programmable Gate Arrays, which developers can customize, and they are well-suited for compute-intensive applications.
- AWS X-Ray, a tool to analyze and debug distributed production applications, was made generally available, along with Lambda integration. AWS will start to charge for usage May 1, with pricing based on the number of traces recorded and analyzed. Users can record 100,000 traces and retrieve or scan 1,000,000 traces per month for free.
- Amazon Aurora support for PostgreSQL was put into public preview in the U.S. East (Northern Virginia) Region. AWS claims the popular open source alternative to Oracle's database provides up to three times the performance of PostgreSQL in traditional environments.
Trevor Jones is a news writer with SearchCloudComputing and SearchAWS. Contact him at firstname.lastname@example.org.
AWS acquisition of Cloud9 pushes it toward IDE functionality
New AWS tools target developer crowd
Developers should be working with these five AWS tools