Storage technologies have been constantly evolving over the past few decades, from dedicated application resources, through server consolidation, to the latest innovations in space saving and performance such as thin provisioning, caching using SSD or RAM, and data deduplication. This technological development has lent itself to the requirements of large data center environments, yet largely ignored the challenges faced by organizations deploying storage at branch or remote locations.
In this five part series we take a look at how storage solutions have evolved and how data storage vendors are finally addressing the challenges faced by distributed organizations in serving their remote sites with cost-effective and highly available shared storage. As mentioned in the last installment, server virtualization and consolidation drive up resource utilization while ensuring that virtual machines only consume the resources that are allocated. This functionality is not only appealing to organizations with data center based server and applications, but to organizations with multiple remote sites. Server virtualization reduces the amount of hardware required at each location lowering capital and operational expenditure.
However, using shared JBOD arrays would introduce a single point of failure and a dedicated storage array would make it prohibitively expensive and complex, especially as the number of sites increases.
To make server virtualization with high availability possible at remote locations an alternative and cost effective storage solution must be used. The introduction of virtual SANs provides the ideal solution.
Storage virtualization using virtual SANs
Virtual SANs are software-defined storage solutions that run as virtual machines, kernel modules or applications on industry standard operating systems or hypervisors, using commodity x86 servers and internal server or directly attached storage. They provide many of the storage features found in dedicated storage arrays, including synchronous mirroring, enabling an exact copy of the data to be replicated on other virtual SAN nodes, providing the all-important highly available shared storage necessary to use the advanced hypervisor features.
Virtual SANs have a flexible architecture and can scale-out from two to thousands of nodes clustered together. This eliminates any single points of failure relating to the storage infrastructure, ensuring application availability is maintained. Compute and storage resources can be scaled-up independently, ensuring that the solution meets the organization’s requirements at all times.
Does this approach address distributed enterprise challenges?
As a software-defined storage solution, virtual SANs are quick to deploy, eliminate server and storage vendor lock-in, giving organizations flexibility and choice over the hardware. By removing the need for a complex and expensive SAN solution, organizations can achieve a smaller IT footprint that is cost effective, simple and quick to deploy making them the ideal storage solution for distributed enterprise.
As shown throughout this series, storage has been constantly evolving, providing more capacity, faster media and a plethora of features to drive up utilization or deliver high availability. Most solutions were aimed at data center environments, not correctly addressing the requirements of remote sites.
The biggest advance for multi-site environments is the development of virtual SANs and software-defined storage. These have effectively brought the remote site infrastructure full circle, allowing storage solutions to be built using commodity servers and internal, direct attached storage, all under the control of sophisticated software, such as StorMagic SvSAN, eliminating the need for dedicated storage arrays.
Virtual SANs have enabled organizations to build simple, cost effective shared storage solutions that deliver high availability required for on-site critical applications.