Tech
Paid Content : This paid content was written and produced by RV Studios of Red Ventures' marketing unit in collaboration with the sponsor and is not part of ZDNET's Editorial Content.

Redefining enterprise storage with OpenStack

Enterprise storage has been an exclusive, niche industry for many years, however new approaches to cloud-based storage is turning this old model on its head.

The lofty world of enterprise storage has been an exclusive, niche industry for many years. Traditional solutions to corporate data management and protection typically involved purchasing large, monolithic storage appliances offered by only a handful of vendors. These centralized, fault-tolerant devices boasted a wealth of software functionality and offered impressive hardware performance and availability guaranties but came with equality jaw-dropping price tags. The emergence of cloud computing (and particularly OpenStack) as a dominant technology platform in the enterprise has introduced a new way of thinking about storage that is turning this old model on its head.

As a component of Intel's software-defined infrastructure initiative, software-defined storage (SDS) is a concept prescribing the separation of physical storage hardware from the software controlling the storage infrastructure. Cloud platforms such as OpenStack simplify the implementation of SDS by abstracting storage hardware and providing access to these resources through a virtualisation layer.

In this series of articles we'll first explore the business benefits of software-defined storage, take a look at OpenStack components designed to implement SDS models, and highlight several open-source storage backend technologies bringing enterprise storage capabilities to OpenStack clouds.

Advantages of software-defined storage

Businesses adopting software-defined storage strategies can achieve significant cost savings through increased storage efficiencies. Such strategies allow consolidation of enterprise-wide heterogeneous storage hardware into a virtual storage pool behind a single abstraction layer, allowing the cloud operator to define policies specifying how and where data should be distributed. This ultimately leads to higher asset utilization and an overall reduction in the number of physical storage devices needed to be purchased.

Traditional storage devices are often encumbered with vendor-specific management tools and proprietary APIs. This problem is compounded when employing multiple hardware units from different manufacturers. By abstracting physical storage into a single logical layer, software-defined storage reduces this complexity. A single unified control point and REST-based API allows centralized management, monitoring, and orchestration of storage across all supported storage devices. This approach also enables capabilities of the virtual storage platform to be extended through a common API interface.

Software-defined storage addresses the physical constraints limiting traditional storage hardware and associated provisioning and delivery challenges. By aggregating existing and new data silos into a single virtual pool, it provides the possibility of limitless storage expansion regardless of physical location. This aggregated storage pool can divided up and allocated based on pre-defined policies to automate the provisioning process.

Lastly, software-defined storage simplifies the management and flow of disparate data type within an enterprise. There are generally three distinct data storage mechanisms: block-storage for compute nodes, file-storage for documents and images, and object storage for unstructured data. It's quite common for business workflows to transmit data from one storage mechanism to another. By separating data access from physical storage, SDS allows this data to be manipulated without moving it between different storage devices, further improving storage and cost efficiency. By reducing costs associated with dedicated storage, SDS makes storage-intensive operations such as data warehousing and analytics more accessible and affordable for businesses.

Storage Acceleration

Intel Intelligent Storage Acceleration Library Project (ISA-L) assists software-defined storage technology by providing a collection of low-level storage functions that can be used in storage applications. Taking advantage of optimizations found in particular Intel Xeon CPUs, ISA-L can be used to increase performance and reduce storage space requirements of SDS solutions delivered in OpenStack clouds.

Summary

Software-defined storage is becoming an increasingly important and widely discussed topic. The benefits it brings to the enterprise are many, including increased storage efficiency, greater flexibility and larger cost savings. The OpenStack platform makes it easy to implement and deploy SDS in an enterprise cloud, and in the next article we'll outline how SDS is architected within an OpenStack cloud.

Editorial standards