| By Maureen O'Gara |
Article Rating: |
|
| January 9, 2013 09:00 AM EST |
Reads: |
2,320 |
Amazon Web Service has fielded what it calls High Storage instances, a new EC2 instance family for applications requiring fast access to large amounts of data.
These new instances are currently available as a single instance type, Eight Extra Large (hs1.8xlarge) and provide customers with 35 EC2 Compute Units (ECUs) of compute capacity, 117GiB (gibibytes) of RAM and 48TB of storage across 24 hard disk drives.
The widgetry is capable of delivering more than 2.4GB a second of sequential I/O performance.
With large amounts of direct-attached storage per instance, these High Storage instances are good for data-intensive apps like Hadoop workloads, log processing, data warehousing and parallel file systems to process and analyze large data sets in the AWS cloud.
Amazon says these new instances can power Redshift, its new petabyte-scale data warehousing service and will be important for customers using its Elastic MapReduce to process large quantities of data.
The widgetry is supposed to lower the cost of processing large data sets with Elastic MapReduce.
According to MapR, Hadoop distributions offer enterprise-grade Hadoop features such as high availability, data snapshotting, mirroring across availability zones and NFS mounts. High Storage instances are ideal for these distributions.
Users can launch High Storage instances using the AWS Management Console, EC2 and Elastic MapReduce Command Line Interface, AWS SDKs and third-party libraries.
High Storage instances are currently available in the North Virginia region and will be available in other regions.
Customers can purchase High Storage instances as On-Demand or Reserved instances.
Detailed pricing information is available at the EC2 pricing page.
