Here comes Object Storage

Image of connected objects

Posted on Wednesday, December 16, 2015

The largest benefit of big data is also its greatest weakness. Having almost unlimited information to support analysis is invaluable, but those insights are only valuable when delivered in a timely manner. The more data your business stores, the more there is to process, increasing the time it takes to unlock insights.

In the age of low-cost storage, adding capacity to support a big data is the least of your business’ problems. Instead you need to find a way to streamline access and processing of the information collected.

Object storage – data with your data

The answer to this problem, as defined by some vendors including EMC, is the use of ‘object storage’. Essentially this approach creates metadata records for each item of unstructured data, helping to identify and classify the information.

Done correctly, metadata can significantly reduce the time and processing power required for big data analysis. Rather than searching the contents of the files themselves, object storage allows data analysts to query the indexed metadata, returning results far more quickly than traditional scale-out storage systems.

Systems like EMC’s Atmos platform provide the necessary scalability for big data storage, along with automated metadata indexing to streamline the process. These object-based storage devices (OBSD) can also automate common functions helping to streamline data management based on metadata rules you define.

A decision that needs to be taken now

Although you can begin collecting information for big data analysis immediately, there needs to be a strategy in place for how you will perform analysis in future too. Fortunately EMC customers have it slightly easier, being able to integrate Atmos into their existing storage infrastructure.

This not only allows them to launch their big data programs immediately, but also helps them push older storage arrays back into service. Under the data object regime, original files can be stored on older, slower disk arrays, whilst metadata is accessed from speed-optimized systems for time-critical analysis, and maximizing returns on existing hardware investments.

big data, and the insights your business extracts from it, will become the difference between you and your competitors. Contact CDS today and let us help you clarify your big data strategy by repurposing post-warranty hardware for scalable storage.