News, Press, & Events

File-area networks: the key to managing data growth

January 31, 2008 - Nigel Burmeister, F5 Networks

Here’s a common scenario: a financial services company experiencing rapid business growth was adding knowledge workers at a hectic pace and expanding globally as well. The IT department was overburdened as it struggled to keep pace with the increase in data associated with existing applications, the rollout of new applications, the growth in the size of file data, the increasing use of multimedia content, and the regulatory need to retain more data for longer periods of time.

The challenges associated with data growth:

  • How do you control the costs of storing and managing skyrocketing amounts of data?
  • How do you quickly and efficiently provision new capacity to meet business needs?
  • How do you back up and protect the burgeoning amount of data?
  • How do you ensure the latest data is promptly available in remote offices?
  • How do you ensure consistent data access and effective collaboration across geographies?

For this company the solution was a file-area network (FAN). Similar to the way in which a storage-area network (SAN) improves the management of structured data and block storage environments, a FAN helps improve the management of unstructured (file) data and file storage environments. A FAN provides centralized, heterogeneous and enterprisewide network file management and control.

Like a SAN, a FAN potentially comprises many technologies: file virtualization, WAN optimization/WAN acceleration, data classification, global namespace, and information lifecycle management/automated tiering. Unlike a SAN however, a FAN does not require new infrastructure, but rather enhances the network and storage infrastructure that already exists.

One of the biggest problems related to the management of unstructured data is the binding between applications, users and the storage devices themselves. While the challenges associated with data growth necessitate change in the storage infrastructure, this rigid coupling makes it difficult to adapt to those changes.

A FAN solves this dilemma by decoupling access to file data from the physical location of the data. The FAN decoupling layer may comprise several technologies:

  • File virtualization removes the physical-location dependency by presenting a global namespace to users and applications that effectively decouples them from the physical locations of their files. With file virtualization, storage-management operations such as provisioning or migration can be performed at any time without affecting applications and users.
  • WAN optimization removes the geographic-location dependency, eliminating many of the undesirable characteristics of the WAN, such as high latency and limited bandwidth. With WAN optimization, data can be centralized to lower the cost of management and increase control while still maintaining the access performance needed by applications and users.

FANs bring another key capability that improves the management of unstructured data — intelligent services. Intelligent FAN services are policy-driven advanced file controls that perform specific tasks on the stored data. FAN services include:

  • Migration: to move files nondisruptively between heterogeneous storage devices.
  • Replication: to copy files nondisruptively between storage devices and geographies.
  • Placement: to automatically assign files to specific storage devices based on attributes of the given device.
  • Classification: to allow indexing of the data for use with policy-based controls.

These are just some examples of current FAN services; additional services will evolve over time.

FAN services make it possible for organizations to reduce the total cost of ownership (TCO) of their file-based storage through automation, improved file-storage performance and flexibility, and streamlined administration and management of file data.

So, how does the FAN solve the real-world problems experienced by the financial services company mentioned above:

The decoupling layer allows:

  • Additional storage to be provisioned on-demand to meet capacity requirements, without affecting existing applications or users.
  • New storage technologies to be easily integrated to support new applications or media types.
  • Remote offices to have timely access to the latest data.
  • Users to gain consistent access to corporate data regardless of geographic location.

Value-added services provide:

  • Streamlined backups by seamlessly moving static older data out of the primary backup data-set and off of premium storage.
  • A transparent way to place different data types, such as multimedia, on appropriate storage platforms.
  • An automated way to place data onto longer-term retention repositories for compliance reasons, and yet quickly recall that data should the need arise.
  • A means of replicating file data from heterogeneous storage systems locally or remotely in order to protect that data in the event of equipment failure or disaster.

In short, the FAN dramatically improved this organization’s ability to manage its unstructured data environment. The benefits combine to provide a lower total cost of ownership (TCO) for file data, increased user and administrator productivity, and ultimately increased ROI. Organizations that have adopted FAN technologies report significant savings, which typically come from reduced disk expenditures and from the streamlining of backup, migration and disaster-recovery processes.

FANs are making it possible for organizations to finally resolve the paradox of having to build increasingly costly and complex storage infrastructures to meet demand for storage. Using FAN technologies, organizations are able to satisfy the growing demand for storage and improve storage performance while reducing the storage TCO.

Burmeister is senior product marketing manager, Data Solutions, at F5 Networks.He can be reached at n.burmeister@f5.com.