Amazon Speeds Analytics With High-Memory Compute Instances

Amazon Web Services (AWS) on Tuesday announced the availability of high-memory cluster instances on the Amazon Elastic Compute Cloud (EC2). The workload-specific cloud services are aimed at memory-intensive uses such as in-memory analytics and databases, caching and scientific computing.

"Real-time applications used by healthcare providers, social networking companies and advertising technology providers require large amounts of memory to maintain high-performance," said Peter DeSantis, VP of compute services at AWS.

Amazon cited customers SAP and Engine Yard as users of the high-memory cluster instances. Engine Yard said it will use the instances to support business analytics, advertising and social-media-analysis applications that its customers are building in the cloud. SAP offers its Hana in-memory database as a service on EC2, and the new instances are a perfect fit, according to Sapan Panigrahi, VP of Hana Cloud.

[ Want more on Amazon big data services? Read Amazon Redshift Leaves On-Premises Opening, Says ParAccel. ]

"The new High Memory Cluster instances provide sizeable CPU cores and storage to run even larger workloads," Panigrahi said in a statement.

[Editor's note: Adding pricing detail, on-demand rates for High-Memory instances range from 45 cents per hour for "extra large" to $2.28 per hour for "quadruple extra large" instances. High-memory cluster instances are $3.50 to $3.83 per hour.]

The new service is the third in a growing portfolio of workload-specific instances introduced over the last six months. AWS previously released high-storage instances and high I/O instances.

The high-storage instances are geared to customers who use Hadoop, high-scale data warehouses and parallel file systems to process and analyze large data sets in the AWS cloud. High I/O instances are aimed at high-performance clustered databases, particularly NoSQL databases like Cassandra and MongoDB, and they're typically used for media streaming, gaming, mobile apps and social networking.

High-memory instances are available on Linux or Windows and will run on two Intel Xeon E5-2670 processors, two 120-GB solid state drives of instance storage, high bandwidth networking, and 244 GB of RAM. The services can be launched and managed using the AWS Management Console, the AWS and Amazon EC2 Command Line Interfaces, AWS SDKs and third- party libraries.