Satisfying a WAN's need for speed may not be the best and most cost-effective way to improve applications performance...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
and access across a network.
A more constructive approach is to make better use of current resources, use tools that proactively monitor network traffic, and do your best to spot and eliminate issues before they become end-user problems, says a report on "best in class" optimization techniques released by The Aberdeen Group.
For its study, Aberdeen looked at companies in a variety of fields, including defense and education. Roughly 30% of the respondents were from large firms (annual revenues above $1 billion) and 41% from midsized enterprises (annual revenues between $50 million and $1 billion).
The research and final report "Application Delivery over the WAN: Acceleration Alone is not Enough" was underwritten by a cross-section of WAN optimization and performance vendors, including Blue Coat Systems, Cisco Systems, Exinda Inc., Riverbed Technology, and Expand Systems.
Putting WAN best practices in play
As part of the study, Aberdeen surveyed 308 companies with the goal of identifying best practices for managing application performance over a WAN.
To achieve "best in class" status, companies must:
- Deploy multiple application delivery functions, like caching, compression, SSL acceleration, and so on.
- Develop capabilities for evaluating test results before making a purchase decision.
- Deploy a single platform for managing applications performance visibility and quality of service (QoS).
Once best-in-class actions are put into play, companies were able to improve application response speed more than 11 times, the report points out. Some of the top actions taken by companies to improve WAN performance include:
The No. 1 objective, however, was to maximize the utilization of an existing infrastructure or make better use of current resources before adding any more, the report says. Although 85% of the best-in-class companies surveyed reported that their WAN traffic increased over the previous 12 months, only 43% opted for costly upgrades or new appliance purchases.
One of the top concerns in matching the demands of mission-critical applications to WAN capabilities is identifying performance issues before they affect end users, the study notes. Others include an increase in the amount of voice and video traffic, an inability to adequately view the end-user experience, and an increase in the complexity of applications. These findings are consistent with the results of previous studies that pinpoint a need to employ solutions that go beyond traditional WAN optimization techniques like compression and TCP acceleration.
Aberdeen recommends that classic optimization techniques be paired with proactive management of applications performance, which means using tools that constantly analyze network and applications performance. The report concludes that "accelerating network traffic will not be enough to satisfy changing needs of business users."
The WAN report also points to the difficulty most companies -- even those with best-in-class practices in play -- have in developing a strong business case for deploying optimization solutions at remote branch offices. The biggest roadblock is justifying the total cost of ownership (TCO) for hardware solutions at these remote sites, which has become a critical factor when evaluating technology solutions, Aberdeen notes. This could be a major reason why many companies are looking into software acceleration techniques as an alternative to adding expensive hardware appliances.
Other considerations -- besides reducing traffic snarls -- that are key when evaluating WAN performance solutions include: scalability, data security and ease of deployment issues.
A complimentary copy of Aberdeen's WAN optimization and performance report is available until July 31, 2009, from the company's research website.