Amazon Upgrades Performance Of Elastic Block Storage Service As Cloud Wars Heat Up

At Amazon Web Services' re:Invent conference last November, CTO Werner Vogels promised partners in his keynote that storage options for the world's largest cloud would soon get bigger and faster.

On Thursday, AWS delivered on that promise with upgrades to its Amazon Elastic Block Storage (EBS), which complements EC2 compute instances, in both the midlevel and premier tiers. Both EBS options have employed solid state drives since last June.

"With this release, you no longer need to stripe together several smaller volumes in order to run applications requiring large amounts of storage or high performance, including large transactional databases, big data analytics, and log processing systems," Jeff Barr, chief evangelist for Amazon's industry leading cloud, said in a blog post.

Barr said General Purpose users can now provision a storage volume of 16 terabytes that offers 10,000 baseline IOPS and the option to burst into higher input/output performance when the situation demands, such as in a boot sequence.

Sponsored post

[Related: Amazon Cloud Introduces Solid State Drives For Block Storage Volumes]

Amazon has boosted performance of its premier storage offering, Provisioned IOPS, to 20,000 IOPS from the 4,000 of the previous specs while also upgrading from 1 to 16 terabytes of capacity per volume, Barr said.

Barr said the new volumes transfer data more than twice as fast, with maximum throughputs of 160 MBps for General Purpose customers and 320 MBps for those using Provisioned IOPS.

The cloud storage wars are again heating up, with a major shot fired by Google last week on the other end of the storage spectrum.

Google released its Nearline storage offering, which it says dramatically decreases industry-standard access time for data locked in cold storage. According to Google, Nearline retrieves data significantly faster than Amazon's Glacier service.

Todd Helfter, database operations manager at Jersey City, N.J.-based Datapipe, one of Amazon's largest cloud partners, told CRN the new storage capabilities allow the managed cloud provider to deploy larger database environments without having to go through the effort of striping smaller disks into larger ones.

"In an environment that often depends on snapshots for database backups it is significantly easier to snapshot a single disk volume over multiple volumes striped into a larger disk," Helfter told CRN via email.

He added 20,000 IOPS is fast for any disk, especially one as large as 16 TB.

Ali Hussain, CTO of Flux7, a DevOps-focused solution provider and AWS partner based in Austin, Texas, told CRN via email that "larger volumes are a much appreciated feature in the present world of growing data."

Flux7 often has to setup multiple storage volumes for customers in a RAID configuration, which increases the complexity of the system as well as setup and maintenance costs.

"This is especially a pain point when dealing with big data tools like Hadoop and Cassandra," Hussain told CRN.

But upgraded volume sizes mostly end that practice, allowing Flux7 to deliver greater value to customers. The larger EBS volumes also will help customers afford to use Amazon's Relational Database Service, he said.