When a networking problem occurs, the answer is easy. Throw more bandwidth at the problem, right? Unfortunately, that doesn’t always lead to a resolution.

“Many companies experience performance issues that on the surface appear to be bandwidth-related, but may actually be tied to poor application performance, improper network design, or suboptimal bandwidth utilization,” says Michael Hall, a network solutions architect at CDW. “Often, the hasty response to these issues is to add bandwidth, which in many cases does nothing but increase internet cost.”

One example is application slowness on what could be mistaken for a sluggish network. Instead of making the lazy assumption that bandwidth should be added, the issue could be with server performance or something else. Providing more bandwidth will obviously speed up a network, but it might not solve an issue with a router or server. In some cases, purchasing an application server may provide greater gains than adding a fatter pipe. So how do you decide what to do?

What’s The Diagnosis?

The first place to start is with a diagnosis, which should be done using a network analyzer. Take the case of those common bandwidth hogs, VoIP and Unified Communications.

“IT managers anticipate that the addition of VoIP will increase their bandwidth needs and, as a result, automatically upgrade their bandwidth capacity,” says Stephen Brown, product marketing manager at Network Instruments. “The increase in bandwidth may waste money without addressing the issue causing the VoIP problems.”

VoIP, then, isn’t the type of technology that network managers should attempt to implement without doing a little legwork up front. A good network analyzer, a test environment, and a gradual implementation will help SMEs to locate the actual bottlenecks and take appropriate action. By understanding the current bandwidth load and existing network and application demands via a network analysis, you can determine the exact cause of slowdowns and embark upon an effective resolution.

“Long-term application performance monitoring and trending is critical for establishing baselines of normal network performance,” says Brown. “You should rely on your network analyzer daily, rather than pulling it off the shelf only during an emergency.”

Predeployment Assessment

Many people only discover unexpected bottlenecks after the fact. That’s a sign of bad planning and results in such things as CRM (customer relationship management) systems that have absorbed vast amounts of bandwidth yet still remain buggy. The best approach is some form of predeployment assessment using a network analyzer or another type of network tool. This assessment allows you to prepare your network ahead of time for the added traffic without surprises.

“By conducting a predeployment assessment of the network, SMEs will better identify and resolve obstacles that would get in the way during implementation and create performance baselines that can be used to evaluate post-deployment success,” says Brown. “Predeployment assessments present a great opportunity to discuss expectations with managers and plan appropriately for long-term network growth. All of this is much easier to do prior to deployment, rather than after the fact.”

No Need For Probes

A thorough diagnosis of the network prior to buying more bandwidth is key. A proper diagnosis might show any number of issues: misconfiguration, bandwidth abuse, hardware/software failures, or no problem at all.

For those who don’t already have sophisticated monitoring systems in place and are unwilling to pay the price for network probes, there are ways to find out what is going on without breaking the budget.

“There are many different ways to diagnose these issues,” says Raul Duran, field engineer for Plixer,  which makes network management tools. “Netflow or sFlow analyzers offer great visibility into network traffic without the need for probes, so they are easy to implement.”

SMEs, then, don’t need to worry about having to spend a year’s budget for enterprise-level network monitoring tools to be able to diagnose network problems. According to Duran, there are quite a few programs offered free that can do the job. Although the IT administrator might not get all the tools they want from a free version, it’s certainly a start. And, in general, even sophisticated Netflow and sFlow tools are relatively inexpensive compared to enterprise-class products and are supported widely by the vendor community. SNMP is another easy and cheap way to monitor devices attached to the network. It provides plenty of valuable information, has been around for a long time, and is supported in almost all business level network devices.

Using Netflow, for example, Duran gives the example of an SME that noticed unusually high network utilization at a remote location. A Netflow analyzer isolated three IP addresses that were responsible for doubling bandwidth utilization. They were running online backups of their laptops during business hours. Once the problem was found, the company put a stop to this practice immediately.

Optimizing The WAN For Remote Users

Many user complaints are about problems in the field when they tried to connect to the SME’s file server over the Internet, attempted to access the database from hundreds of miles away, or tried to open and run an important business application remotely. If it took ages to open a document, access an application, or view a database record, then the issue might be addressable with WAN acceleration.

Instead of expanding the amount of network presence, this technology seeks to optimize the way the existing network is being used. This is accomplished by such techniques as compressing the traffic that is crossing the WAN, caching and accelerating the applications involved by improving their performance.

“When diagnosing WAN performance issues, many companies find that improving or optimizing the way they use their WAN links can result in a performance increase of 70% or more without adding bandwidth,” says Hall. “WAN optimization products use a number of different strategies, the most common being caching and compression. These WAN optimization products are great for companies with branch offices or mobile users and exhibit a short ROI timeframe due to increases in productivity or reduction in bandwidth requirements.”

Reviews

Leave a Reply

You must be logged in to post a comment.