In an effort to drive data traffic efficiencies across the enterprise, an increasing number of IT departments are deploying WAN optimisation devices.
According to a recent InformationWeek report 64% of businesses have either implemented, or are considering, WAN optimisation products. The key driver for WAN optimisation is, according to 74% of those organisations, ‘improving client/server application performance’. However, organisations must plan the implementation of this technology carefully otherwise there is a real risk that they won’t realise any such benefits.
Simply put, if they invest without first auditing their application landscape to pinpoint the origin and impact of any problems, they are unlikely to find any reward from costly WAN optimisation devices.
Running before you can walk
WAN optimisation technology is designed to speed up the delivery of applications crossing the Wide Area Network. It does this using a range of methods including storing regularly-accessed data in local caches which stops a file crossing the WAN more than once. The theory is that, with less data crossing the WAN, pressure on the network will ease and data will be delivered to the end-user far quicker.
However, as is often the case with much-hyped technology, some businesses have been misguided by the promise of ‘faster speeds’ and ‘reduced costs,’ forging ahead with WAN optimisation and then being frustrated with the results. Due diligence is critical but in relatively new areas of IT such as WAN optimisation, it is understandably difficult to apply. However, in our experience before any investment to improve performance is made, organisations first need a detailed view of current operations. Without this, they’ll never know if the investment was really worth it and they’ll be unlikely to achieve the expected return on investment.
WAN optimisation isn’t a silver bullet. Businesses must first audit their IT infrastructure, looking at everything from the time it takes for data to reach the end-user, to analysing when the network is at its busiest, who is using it and what information they’re accessing. If there are slowdowns they need to understand what’s causing it. Gaining this level of understanding before you commit to investing in WAN optimisation means businesses can figure out exactly what is causing any data slow-down.
They can check whether it is a network problem or whether it’s something less severe – perhaps a minor glitch in the system which can be easily remedied. For example one organisation we worked with realised a security camera was transmitting live images over the WAN when the feed only needed to go a few yards to the security team. Remedying this would cost far less than if the business opted for WAN optimisation which would in this instance have no effect.
Seeing beyond the network
Central to understanding whether WAN optimisation can make, or indeed has made, a difference is seeing what the end-user is experiencing. A lot of WAN optimisation controllers have their own reporting tools which provide network performance data but don’t indicate performance from the perspective of an end-user operation or business process.
In certain cases, the technology may have improved the state of the network but overall made little difference to the end-user. For example, whilst the network performance has improved, an issue outside of the network such as a problem with the user’s PC could mean data is still very slow to reach them. This is where real-time, end-user monitoring proves vital. With this in place, an organisation can see beyond the network to ensure improvements are actually reflected in the end-user experience. If monitoring tools show that performance is perfect all the way up until the final stage of delivery – the likelihood is that any problem is with the end-user’s equipment rather than a wider network issue.
In an effort to analyse the impact on the end-user, some organisations use tactics like stop-watch analysis to see how fast information is reaching the end-user, but this point-in-time measurement does not reflect the impact and dynamics in today’s constantly changing network environment. Taking a snapshot view of performance might show excellent service, but an hour later the view could be very different. To monitor performance of modern IT environments where a whole range of external services such as cloud vendors and hosting providers play a part in data delivery, these ‘point-in-time’ tests are not enough. Faultless end-user performance requires continual, real-time monitoring.
Gauging the impact – three simple steps
In summary, to make the most of WAN optimisation technology, organisations should follow three simple steps.
- Step 1: It is essential that organisations take the time to do rigorous tests to understand the health and performance of their IT infrastructure. To do this effectively, these organisations need to use end-to-end performance tools to assess current service from an end-user perspective. In some cases, it may look like all lights in IT are ‘green’ but what the end-user is experiencing does not reflect this. Ultimately, an organisation will be judged on the service it provides to the end-user. WAN optimisation could aid this but, if there’s no end-to-end view, the businesses will not know a) if the technology is needed or b) how it has improved performance once implemented.
- Step 2: This next step is inherently linked to ensuring the business has an end-user view. With this kind of insight, organisations will be able to pinpoint problems in the application delivery chain, and therefore whether WAN optimisations could help resolve these. This should be done before even thinking about implementing this technology. It could be that a performance problem is down to a glitch or anomaly, rather than something more serious. By doing this initial testing, businesses will be able to pinpoint such issues and potentially save themselves millions of pounds.
- Step 3: Move ahead and implement WAN optimisation technology, providing it is the right option. If the business has gone through the first two steps, it will know whether WAN optimisation is worth the investment and will also have a good idea of exactly what improvements the technology should bring. As with any investment, businesses need to understand what they’re trying to improve upon before spending. Without this, there is no way of knowing whether WAN optimisation was worth the considerable investment. Simply put, you can’t put the cart before the horse.
Michael Allen is director of IT service management EMEA at Compuware
Find your next job with computerworld UK jobs