Block-level storage virtualization, one of two primary types of storage virtualization, is the separation of logical and physical storage so that the logical storage can be accessed regardless of the structure of the physical storage. Block storage virtualization is applied at the storage area network (SAN), whether Fibre Channel-based or iSCSI-based. It pools storage devices -- in its truest form, from multiple vendors -- so that they appear as one repository of storage that can be managed collectively. As a result, IT pros have much more flexibility in the way they manage the storage system.
In this four-part series, Eric Slack examines how to virtualize storage at the block level. Part 1 lays out the main benefits of block virtualization. In parts 2, 3 and 4, Slack addresses how to apply storage virtualization technology, respectively, at the server level, on the storage array and, finally, as a network-based storage virtualization appliances.
Table of contents:
In this introduction to our series, Slack explains the details of block-level storage virtualization, its benefits for IT departments and the reasons behind why many storage pros are beginning to virtualize storage. To start off, he explains that block virtualization provides an adaptable, logical arrangement of storage capacity to both users and applications. It makes it easier for storage administrators to provide storage capacity when and where it’s needed, at the same time allowing behind-the-scenes work -- including expansion, maintenance and protection of data -- to be hidden from end users.
In Part 2, learn how to virtualize storage on a server level. Sometimes referred to as host-level, this type of virtualization technology is applied in a software layer on a logical storage volume manager -- in between the operating system and applications or file systems. These solutions usually virtualize the storage that is attached only directly to the server, though there are some cases where virtualization can affect multiple storage subsystems connected by a SAN. There are a number of virtual storage appliances on the market, and Slack outlines the most notable options for IT pros to review, including the vSphere Storage Appliance, HP LeftHand Virtual SAN Appliance, DataCore SANsymphony-V and FalconStor NSS Virtual Appliance.
When server-based storage virtualization began to evolve, virtualization within storage arrays began to develop. In Part 3 of our storage virtualization series, learn how virtualizing storage on the array has become possible and why it’s important for efficient provisioning and management of storage arrays. Slack explains that when turning to virtualization for multivendor, or heterogeneous, storage consolidation, array-based systems aren’t as common as network-based systems. In this tip, he outlines the Hitachi Data Systems and NetApp solutions -- two of the multivendor array-based solutions for virtualized storage available on today's market.
In Part 4 of his series, Slack evaluates network-based storage virtualization appliances. He explains that some form of storage consolidation is usually a part of the majority of storage virtualization cases, which means that placing the engine of the storage virtualization on the network is a logical decision. Furthermore, he shows readers how network-based implementations can be embedded into switches or run on dedicated commodity hardware, removing the overhead process for the CPU host and the need to purchase new arrays just to get storage virtualization technology. Slack assesses three in-band storage virtualization appliances on the market today: DataCore’s SANsymphony, FalconStor’s Network Storage Server (NSS) and IBM’s SAN Volume Controller (SVC).