This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.
As the WAN matures, with advances in robust applications, cloud delivery and other technologies, optimization tools have had to keep pace. It is no longer just about classic compression and protocol acceleration. The new role is about application performance, delivering everything from application visibility to quality of service (QoS) control. In short, the focus is on true WAN governance.
The proliferation of cloud computing will have a major impact on the WAN and, specifically, on the performance of business-critical applications that cross the WAN. Data centers and companies alike must be prepared to address this shift since, with cloud computing, the WAN traffic matrix becomes much more complicated and stresses the network like never before, requiring movement of massive amounts of data.
Enamored by the promise of the cloud, organizations looking to reduce costs and improve service levels must first prepare the network to support significant increases in bandwidth usage and Web browsing traffic. Organizations must inventory both the critical and noncritical applications running on the network and also assess use-case scenarios for events that may tax the WAN beyond its normal capacity.
There are two ways to address the need for greater bandwidth: Purchase more or use what you have more effectively through WAN optimization tools, which monitor and regulate the flow of applications across networks. Adding more bandwidth is a stop-gap solution, as more is rarely enough in the long term. Critical applications will take as much bandwidth as possible because of the way that the Internet protocol is designed.
TCP (Transport Control Protocol) operates as a layer for protocol on top of IP. Designed to be able to share limited, poor-quality resources, TCP is structured to seize available bandwidth. When the network is struggling to deliver resources, TCP will sense delays and slow down, creating congestion situations. It's a protocol where every data transfer, every application, will try to take as much network resource as possible. This means more bandwidth, applied indiscriminately, isn't the solution.
The solution lies in having tools that allow an IT department to control applications as they flow across the network. With WAN governance, IT departments can see how applications are moving, who is using them and for what purpose. IT can then prioritize applications accordingly.
True WAN governance is about aligning business priorities and IT priorities over your global WAN. It means being able to identify business-critical applications on the network and their level of performance. IT needs to ensure user productivity and satisfaction and also provide the measurement and reporting on actual delivered application performance, in a way that is easy to understand. With WAN governance, enterprises can finally manage the cost/performance tradeoffs at the IT level (i.e., rightsizing your network according to application performance SLAs).
For today's large enterprise, what's most important is to handle the application performance problem strategically and in a global manner, with a structured approach, not on a site-per-site or problem-per-problem basis, involving the global network and the entire application portfolio.
The ultimate objectives of application performance are threefold: to address the issue of productivity through guaranteed performance of business-critical applications; to ensure end-user satisfaction and reinforce the positive image of IT within the enterprise; and to minimize or reduce costs in delivering application performance to end users. WAN optimization is one of the techniques that can be used to reach these objectives.
Application delivery services are also critical for smaller companies that are rapidly adopting cloud, which means that access to the network is vital for their business. Yet until recently, the SMB marketplace was put off by the high price of application performance services, as well as by their complexity. New application performance services are both more affordable and easier to use, enabling these companies to support traditional portfolios (ERP, CRM, etc.) as well as cloud-based applications.
But regardless of the size of his organization, when should IT seek an application performance services? The most obvious is when they experience performance problems in lines of business, as well as by end users; but also when they implement new IT projects, such as cloud computing, or make significant changes in the IT environment through new applications (SaaS, UCC, etc.). But even when there are no problems or major IT changes taking place, it's a good idea for companies to think strategically about its IT governance and application SLAs on a regular basis.
It is very critical that application performance services apply to the entire suite of applications. We see more and more bandwidth being used by bandwidth-hungry applications combined with a mix of recreational and business applications on the network. It's critical for an enterprise to be efficient, and application performance services should be able to provide global traffic control over the network.
IT need good visibility to what's going on in the network. In many cases, IT managers discover applications on their networks that they never knew existed. Features such as per-user control provide dynamic protection of application performance against unexpected behavior.
Some application vendors have already made efforts to build in WAN governance. But enterprises cannot solve application performance problems from the application point of view only, because they have to handle a combination of all flows in a dynamic way. True WAN governance lies not with application vendors, but with corporate IT, which can see the whole picture.
Read more about lan and wan in Network World's LAN & WAN section.