|
Executive Summary
As businesses become increasingly dependent on the effective delivery of IT services over far-flung enterprise networks, technologies that optimize the performance of applications over the WAN have become increasingly attractive. In fact, this type of technology may be indispensable for companies operating on a global scale.
There are, however, many types of WAN optimization and application performance acceleration solutions from many different vendors. And IT budgets are not unlimited. So any decision-maker contemplating the acquisition of such a solution must somehow select the one that will deliver the greatest value, based on the company’s specific IT environment and specific business requirements.
This white paper explains how to best make this critical technology decision. It provides an overview of best practices for defining requirements and comparing alternative solutions. Most important, it offers a metric – performance ROI – that IT organizations can use to select the WAN optimization solution that delivers the greatest business value.
Shunra enables these best practices by providing evaluation teams with a controlled benchmarking environment they can use to accurately compare solutions from various vendors. This benchmarking environment allows performance gains at the end-user desktop to be measured for both current and projected network conditions.
WAN optimization can ensure that remote users around the world have ready access to critical applications and services. But it’s important to acquire the right solution for the job. This white paper outlines the principles and practices that will ensure that you do so.
The WAN Optimization/Application Acceleration Solutions Market
Advances in WAN optimization and application acceleration have led to a proliferation of solutions, presenting potential buyers with both greater choice and greater confusion. In fact, vendors marketing these solutions use a variety of names to describe them – including “WAN acceleration,” “application acceleration” and “wide-area data services.” For the purposes of this paper, all of these diverse technologies will be referred to by the single, common term “WAN optimization.”
The following is an overview of the types of application acceleration and WAN optimization technologies available today:
Data Reduction is a collection of techniques aimed at reducing the amount of traffic that actually traverses the WAN. Implemented differently by various vendors, these techniques typically include data compression, caching, and terminal services – as well as recently introduced network memory solutions where only incremental changes in data are fetched across the network, while duplicate requests for the same data are served locally. These techniques are typically more effective for applications with static data and/or where the same data is accessed multiple times.
Latency mitigation applies various algorithms to overcome the adverse impact that network delay has on application throughput and performance. They can include algorithms based on the network layer (such as TCP and CIFS acceleration) and application-specific algorithms for SQL, HTTP, Microsoft Exchange, SAP and/or Oracle, etc. These algorithms typically deliver greater benefits in environments where application performance is constrained by network latency, rather than by bandwidth.
Bandwidth management and QoS prioritize certain network traffic to overcome the performance problems that occur when WAN links are over-subscribed. This way, traffic such as Internet browsing that is deemed less important – or where degradations in performance are considered more tolerable – takes a “back seat” to critical services such as transaction processing. These techniques are typically most appealing where a small number of high-priority applications must maintain their performance even under extreme network saturation conditions, while the performance of other applications can be allowed to deteriorate.
Most commercial WAN optimization solutions combine more than one of these techniques. It is the variety in these combinations – and the associated variation in their impact on different types of traffic – that make it difficult for decision-makers to determine where to best spend their limited budgets.
Performance ROI: A Business-Driven Decision Metric
Faced with such diverse choices, IT decision-makers must carefully evaluate alternative solutions. A technical benchmark is insufficient for such an evaluation, because WAN optimization solutions vary greatly in cost and impact different applications in different ways. By the same token, it is difficult to calculate true ROI, since that would require assigning a concrete value to each second of reduction in an application’s response time – which even if possible is not very practical.
|
