The new world of NAS virtualization: the next generation could fulfill NAS' promise - Storage Networking - network attached storage

Computer Technology Review, Nov, 2002 by Mark Buczynski

Storage virtualization has been a key enabling technology for storage area networks (SANs). Virtualizing storage allows administrators to treat all available storage devices as a general pool of storage, which is presented to the application server as one big disk that can be expanded or modified dynamically without impacting storage availability to applications.

Virtualization is now playing a key role in a new generation of network attached storage (NAS) technology. This new NAS architecture breaks the limitations of traditional NAS architectures and allows administrators to implement global file systems across servers, allows creation of application-specific storage pools that precisely provision storage assets to match business requirements, and enables management of all of these implementations from a single management console easily and efficiently.

One of the limiting factors of NAS has been the inability of a NAS file system to scale beyond an individual server, introducing additional costs and complexity when storage capacity needs to be expanded or re-allocated. The new-generation NAS architecture is based on a distributed file system that incorporates many virtual features into the base system architecture to provide a complete solution to the scaling, performance, and management problems that have plagued the NAS industry.

The use of a distributed file system enables clustering of NAS servers into a single storage resource. This new architecture then implements and expands the storage virtualization paradigm by employing a series of virtual NAS functions. In the SAN arena, multiple virtualization schemes are available--software-based, hardware-based, appliance-based--each with specific strengths and weaknesses. The new NAS virtualization architecture rolls the best of all virtualization approaches into one NAS server by offering not just basic virtualization, but a more robust virtualization implementation that encompasses the entire realm of storage management. This new architecture includes virtual file system, virtual server, virtual interface, and storage pool components all designed to ease the burden of multi-server NAS management.

These innovations provide unprecedented NAS functionality, including:

* Transparent multi-server scaling of capacity and performance.

* Multi-server scaling of capacity and performance of a single file system.

* Global file system across multiple servers.

* Application-specific storage pools.

* Non-disruptive, online data movement.

* Single management console and management view across servers.

* Segregation of clients for secure resource sharing.

* Segregation of clients for logical management.

* Transparent client access port failover.

* Non-disruptive client access load distribution.

Virtual File System

The virtual file system (VFS) can be thought of as the universal storage container of the NAS storage system. A VFS can consist of a directory tree and associated files that can be viewed and manipulated as a singular storage container. A VFS may be assigned to a user, group of users or applications. VFSs have quotas and access permissions. Clients access files as they normally would using a standard NAS server using with? shares or mount points. The client files are stored in VFSs on the NAS server. The NAS server distributed file system takes responsibility for virtualizing or mapping the clients' actual shares or mount points to a particular VFS within the storage system. The distributed file system provides a de-coupling of the actual user's share or name space from the physical location of where the data or VFS resides. The users are unaware of the actual physical location of their data. It may be on a local NAS server or a NAS server in a different city. The virtual file system plays a big part in elimi nating location dependency and downtime required for reconfiguration and scaling because the file access has been de-coupled from the physical storage location.

Virtual Server

The architecture logically groups select storage resources into a single virtual server (VS) that can be assigned and relegated to a particular user community. Typically a virtual server would be configured with client access ports, VFSs, and an administrator. Virtual servers can span multiple physical servers that may be geographically dispersed. The virtual server handles the presentation of storage resources as a combined whole to the clients. Virtual servers provide segmentation and logical isolation of users on a shared set of resources, securely and transparently. It works by associating each virtual server with a set of network ports and a set of virtual file systems. Only requests from ports configured on a virtual server may access VFSs stored on the virtual server. Users only see VFSs associated with their respective virtual server. A virtual server may have one or more administrators, allowing multiple levels of storage management.

Virtual Interface

The virtual interface (VI) provides client access to a virtual server. A virtual interface is mapped to one physical interface on the NAS server. The virtual interface is then assigned to a specific user or user community. A virtual server may have one or more virtual interfaces assigned. Since a virtual server may span multiple NAS servers in a cluster, a single virtual server may have many virtual interfaces assigned on multiple NAS servers. In the event of a failure of a particular physical port, all virtual interfaces may be automatically and transparently failed over to partner virtual interfaces. For load balancing and load distribution, a single virtual interface may be migrated to another physical interface on the same or a different NAS server without user disruption. It is a straightforward but very important part of the NAS server virtual offering because it helps preserve performance, non-disruptive client access and support high-availability configurations.