The world is generating, storing, analysing, transforming, and leveraging more data per year than in the previous ten years combined - and it is expected to more than double each year going forward. At this rate, according to industry analyst IDC, digital data will surpass 163 zettabytes by 2025 (source: IDC Data Age 2025), which would require 163 billion, one terabyte (1TB) drives, to store all of it.
Today's datacentre is rapidly changing as enterprise clients now interact with it well beyond the server room - whether on-premises or in the Cloud. Clients now connect to the Cloud in ways that previously didn't exist as mobile devices have changed the way we communicate both professionally and personally. Not only is data being generated by PCs and smartphones, but also from cars, robots, drones, surveillance systems, sensors, medical devices, wearables, etc. Simultaneously, data associated with photos, video/audio streaming and social networking has grown exponentially.
When we analyse these volumes of data and extract value, our lives are enriched with insight that leads to better decision-making. But not all data has the same intrinsic value, and as such, the cost-effective approach to storing it is not always high-performance primary storage, but sometimes a performance-adequate solution with extreme scalability, advanced availability and durability, and simplified management, may be the better choice.
Introducing Object-Based Storage (OBS) - an alternative storage solution that cost-effectively delivers data at scale and is replacing the traditional file-based Network-Attached Storage (NAS) architectures used widely in today's datacentres. The second part of this article will explore the key features associated with OBS.
OBS is an architecture that manages data as objects as opposed to traditional block- or file-based approaches, and is a viable option for storing unstructured data at petabyte scale. Unlike file-based storage that manages data in a folder hierarchy, or block-based storage that manages disk sectors collectively as blocks, OBS platforms manage data as an object.
Network-Attached Storage is file-based, with data stored inside a folder/path (hierarchical storage) and must be traversed each time that data needs to be accessed. Storage Area Networks (SANs) are block-based in which a collection of disk sectors is logically grouped into a block with a unique address. In either case, there is little or no information about the data stored that can help simplify manageability or support ever-increasing amounts of data (or what is referred to as data at scale).
With OBS, all the data, whether it's a document, audio or video file, image or photo, or other unstructured data, is stored as a single object. Metadata is also associated with the object and provides descriptive information about the object and the data itself. This eliminates the need for a hierarchical structure and simplifies access by placing everything in a flat address space (or single namespace). The unique identifier assigned to each object makes it easier to index and retrieve the data or find a specific object such as a video or photo. Since metadata is defined by users, when leveraged appropriately, enables data analytics or other information discovery techniques for a large volume of data at scale.
Key Features of OBS
While this article explores OBS as an alternative storage solution and its role in cost-effectively delivering data at scale, as more OBS platforms are deployed, replacing traditional NAS file-based architectures in today's datacentres, IT managers are discovering other key benefits associated with OBS, including its extreme scalability, advanced data availability and durability, and simplified data management, all of which will be covered in part two.
Erik Ottem is director of product marketing datacenter systems at Western Digital