Data growth seems to know no bounds
Despite those cynics in the early ‘noughties’ who believed that data growth was just something cooked up by vendors to sell more storage – data growth, particularly un-structured data has continued to grow at a rate that few of us could have ever imagined. Organizations, not only have to care about storing and protecting this mass of data, but also have adhere to an increasing number of regulations which stipulate that they must secure their data, for what in some cases may be literally tens of years, which inevitably means that the data will need to outlive the underlying technology that was originally used to create and store it.
Add in a multiplicity of applications, propriety platforms, private and public clouds and dispersed geographies and the issue of how to store and provide access to this data becomes even more of a headache, particularly in a world where people’s tolerance levels to receiving a response from something like a website is measured in 2-3 seconds.
Ta Da! Object Based Storage (OBS) to the rescue
Like many things in IT, OBS has been around commercially since the early 2000’s, generally being used for specialist applications such as medical imaging and legal archives. More than a decade on and gazillions of data later, organizations are now starting to wake up to the full potential of object storage as a way to be able to store and navigate literally billions of ‘objects’. Unlike hierarchical file-based storage, with OBS each piece of data or record (object) contains all of the information (metadata) that is needed to describe it. Each object is unique and due to its self-describing nature is effectively platform and location independent.
A simple analogy that can be used to help explain OBS is valet parking at the airport, or some other location. When you check your car in you are given a ticket which is later used to get your car back. In between times however, your car could have been moved (hopefully carefully…) countless times without your knowledge, but the only thing that really matters is that you quickly get your car back when you present the ticket. With OBS, the metadata used to describe the object effectively equates to the ticket.
NetApp StorageGRID Webscale
To address the growing object storage trend, we at NetApp have our [link] StorageGRID Webscale solution. Now in its tenth generation, StorageGRID is a mature solution used by organizations all over the world and forms an important component as part of our [link] Data Fabric vision. Depending on your preference, the solution can be provided either as [link] software, or as a [link] packaged appliance using our E-series storage hardware platform. Scalability is massive with up to 70 Petabytes, 100 billion of objects and a geo-dispersed, erasure coding-based architecture across 16 global locations.
And, because StorageGRID uses the language of the Web – Hypertext Transfer Protocol (HTTP) – it’s capable of describing the location of stored data anywhere in the world, making it accessible from anywhere in the world. In other words, a stateless connection to the storage allows the creation of cloud-like applications. Another feature of StorageGRID is that it supports cloud, disk and tape to cater for different cost and availability requirements.
At our Insight tech conference last week in Berlin, NetApp announced the latest 10.2 release which now adds support for NFS & CIFS protocols, OpenStack Swift and RESTful management APIs. Here’s a [link] to the press release which also includes the latest news for: SnapCenter, NetApp Private Storage (NPS), NetApp Professional Services, and CommVault IntelliSnap.
I think that we are going to hear a lot more about object based storage over the coming period, as it starts to become a relevant and essential tool in dealing with the ever-growing volumes and individual instances of data. Think of OBS as a way to address an increasing number of workloads such as: cloud content, big data analytics, image stores, the Internet of things, etc., at a price point that’s difficult to match …
From the EMEA Product & Solutions Marketing team
Links to other associated blogs:
Basic File System Access to StorageGRI?D via s3fs – http://community.netapp.com/t5/Technology/Basic-File-System-Access-to-StorageGRID-via-s3fs/ba-p/110126
Backup – still high on the ‘to do list’ (AltaVault) – http://community.netapp.com/t5/Pan-EMEA/Backup-still-high-on-the-to-do-list/ba-p/112703