SDS redefines storage for demands of future data centres

James Hayes

Wednesday 16 May 2018

As hyper-convergence sets the data centre change agenda and next-generation solutions apply the ‘software-defined’ prefix throughout the infrastructure, IT decision-makers are working through an era of ‘reinventive innovation’. But as a Lenovo briefing explains, IT strategists believe many data centres still do not have the infrastructure necessary to meet the challenges of enterprise digitisation and digital transformation.

The most promising model for improving the dynamism and agility of data centre server, storage and network capabilities is the software-defined infrastructure (SDI) concept integral to the software-defined data centre (SDDC). Facilities whose progress strategy is oriented toward SDDCs should know that, ultimately, they must prepare for an all-or-nothing transformation to achieve targeted ROI. This means that every aspect of data centre infrastructure – from compute power and memory to networks and storage – must be re-specified for a hyper-converged infrastructure that assures high-quality user experience.

The term ‘software-defined’ can be confusing when the impact on data centre hardware is taken into consideration, especially with regard to provisioning software-defined storage (SDS) solutions. Although primarily a hardware exemplar, another of Lenovo’s unique selling points is its commitment to technology partnerships with leading software vendors. The goal is that its products can be delivered custom-configured for optimal performance, as is the case with its SDS solutions.

Enterprise storage fundamentals are being reshaped by three factors:

  • Data sets continue to grow in size exponentially, and ways must be developed to manage these massive datasets before they grow beyond effective control, and business is impacted. The most scalable and expensive SDS solutions are not enough. According to the latest Interop ITX State of Infrastructure survey, the single greatest factor driving change in IT infrastructure is growth of data and data storage technology: 40 per cent of respondents cited growth of storage/data as the factors that are driving the most change in their IT infrastructure.
  • The data that enterprises generate and require is now more unstructured and diffuse. Storage systems must be designed to support a range of data structures, possibly including legacy storage, otherwise they will inhibit business success.
  • Emergent compute-intensive applications necessary to push business forward and maintain competitive advantage, such as real-time data analytics and business intelligence tools, make high demands on all aspects of IT infrastructure, not just the core processors. These applications need to reference data from multiple diverse datasets simultaneously, and having to find and extract stored data across different platforms can add critical microseconds to the process.

Impact of virtualised infrastructures

 Storage management has long evolved beyond the simple act of providing secure and discoverable places to store data assets. Now storage must be optimised for emergent applications that make new kinds of demands on its ability to organise and retrieve data. Technologies such as VMware’s, for example, manage resources dynamically in accordance with an organisation’s workloads. With SDS added into the mix, data centre managers can better organise storage to meet workload and performance requirements in virtualised environments.

Use of SDS is becoming customary – if not essential – with VMware vSAN solutions and open-source solutions such as OpenStack or Red Hat Ceph Storage. Using these technologies can enable data centre infrastructures to become increasingly scalable and, above all, easier to administer. SDS can also ensure direct integration of functions such as backup, restore, compression, and merge and purge.

But there are additional business drivers helping to draw interest in SDS technology, as a report from DataCore reveals. They include simplified management of different storage products and futureproofing of IT infrastructure.

 Open to change

SDS also addresses an issue that besets many organisations: storage silos. Keeping data segregated and not readily accessible is now unacceptable for many advanced line-of-business and decision-support applications. Storage silos can also leave storage capacities underutilised while additional spend on storage array products occurs elsewhere on an enterprise system.

SDS enables a more open and inclusive regimen where storage can be organised to be shared more efficiently across different environments and applications, reducing the need to purchase extra capacity, which keeps CAPEX down. At Lenovo customer Federal Hill Solutions, for example, storage capacity is pooled across nodes.

SDS management has evolved into a specialist technological discipline, and as a total solutions vendor, Lenovo partners with software leaders so that its hardware can be purchased ready-optimised to the software solution its customers want. So Lenovo’s Intel Xeon-powered StorSelect DX8200 product range is designed for use with three leading SDS platform providers – Cloudian (DX8200C), DataCore (DX8200D) and Nexenta (DX8200N) – with each solution aimed at different operational needs.

Lenovo is no newcomer to the software-defined technology. In his 2016 review of SDS, ESG Senior Analyst Scott Sinclair described Lenovo’s leadership in this sector as focused on “architecting solutions specifically designed for [the] software-defined storage era…[Its] solutions retain many of the benefits offered by SDS, while automating the hardware integration”.

Adoption of the Lenovo StorSelect platforms is an investment also in the future of what might be termed ‘storage economics’. Clearly, because it is designed to facilitate speed, Lenovo’s SDS solutions make a quantifiable boost to business success with improved levels of storage performance.

Building the next-gen data centre

Where traditional and web-scale apps co-exist