Objectfying Big Data Storage

Storing a large amount of data is relatively easy, but retrieving it once it is in storage for rapid use in analytics applications can present something of a challenge.

This week we caught up with NetApp, one of the larger data storage and management companies, to discuss how users can address this seemingly simple problem via the introduction of object oriented storage.

It’s probably going to come as little surprise that NetApp thinks their storage offering targeting these big data storage and movement challenges can be addressed with their StorageGRID 9.0 release, but they did make something of a compelling case. According to Richard Treadway who handles the messaging around their big data approach object-oriented storage is the wave of the future for petabyte and terabyte-sized data sets.

Object oriented storage is part of what NetApp is calling Cloud Data Management Interface (CDMI), the main emphasis of the release of StorageGRID 9.0. The idea, more than being able to store a large amunt of data, is being able to go through petabytes of data “We have a lot of customers in the multi petabyte space, we have sixteen customers who have 50 petabyte plus space that are using meta technology. As a result of CDMI, our partners can develop products that can seamlessly plug into our infrastructure.”

Following traditional methods, it can be time-consuming and expensive to recall data stored in a massive data repository, especially if in the multi petabyte range. In simple terms, according to Clifton, this new technology will help drive down costs and increase customers’ business agility by reducing the time and memory required to fetch data. In technical terms, it is the object based CDMI, which labels each file uniquely.

“Object storage is adding a third dimension, you have objects to create to carry the content of the file, you have unique identifiers that allow you to address that object, and then you have metadata that describes the data in the object. So you’re going from navigating a tree to get to a file name over to simply being able to search for the metadata that is in that file.”

Other improvements through StorageGRID 9.0 include increased storage density, industry leading according to Clifton, and larger repositories for unstructured data. But the emphasis remains on the new CDMI which, according to their press release, “enables the creation, retrieval, management, and deletion of data in the cloud.”

“Our distributed content repository solution,” says Clifton “is based on restoring the grid. We have a very scalable and flexible infrastructure that maps the entire environment.”

NetApp has already established itself as a top five data storage and management company. This new release would be another step forward in improving the end-user’s relationship with big data in making search queries faster and less memory-intensive. It also would lessen the memory and cost load on businesses with big data needs.

Related Stories

A Revolution for Open Source Big Data

Hollywood Sharpens Focus on Storage

Versant Throws Magic Cube at Big Data