In an ideal world you would be able to detect and fix network service issues before users noticed there was a problem, says Riverbed Technical Director, Paul Griffiths.
That ideal world is actually a reality with the latest technology. But when looking for that nirvana there are five key questions you need to ask for assessing your options for detecting – and fixing – app and network performance issues.
Firstly, the ground work: How many business-critical applications do you have and how quickly do you need to troubleshoot them when they’re slow?
Determining the types and number of business-critical applications your business uses every day can help you to narrow your focus when selecting a performance management solution. Focus on how fast you need to detect and fix network and application problems – even before they impact your users.
Just finding all the applications running RIGHT NOW across you network can often lead to some unexpected discoveries and insights. For example, shadow IT apps that your users have downloaded and are being used on an ad hoc basis could be consuming more bandwidth than you anticipated. The latest technologies, like SteelCentral network performance management (NPM) and SteelHead application acceleration, can help by quickly and automatically identifying all the applications on your network – business and recreational – so you can effectively troubleshoot, plan for transformation, and police the network, if necessary.
Secondly: Can you guarantee performance of business-critical applications?
Business-critical applications that are being crowded out by a couple of individuals in a branch office listening to streaming radio can cause a slowdown that neither you nor your users can afford. You caught and stopped them this time, but how do you prevent this from happening again? If you are using SteelHeads to optimise the branch traffic, you could implement QoS to make sure latency-sensitive apps like VoIP and video get top priority, business-critical apps come next, and recreational apps are delegated to the ‘best-effort’ category. But is that enough? How do you make sure that QoS is working as expected – that your VoIP phones aren’t experiencing jitter because someone accidently mismarked the local router? Centralised reporting and monitoring can ensure proper categorisation and easy troubleshooting.
Thirdly: Do you have the in-house resources to devote to day-to-day operation of business-critical applications?
Balancing strategic projects with day-to-day operational needs is challenging at the best of times. Throw in a few ‘fires’ to put out and the best laid plans go by the wayside. Establishing your solution requirements for troubleshooting and fixing application slowdowns and outages—for example, focusing on gaining comprehensive visibility of your most business-critical applications—is crucial in ensuring that you get the best fit for your business goals. Be sure to think up front about factoring in any needed implementation services, as well as ongoing maintenance and other support.
Fourth: What will your organisation need from the network in order to handle the challenges now facing the sector?
With digital transformation an essential goal, networks need to provide a platform for the future - enabling rapid adoption of cloud and SaaS for example. Therefore, as with any solution that you plan to keep for a while, you should plan not only for what you need now, but also what you need in the foreseeable future. A good place to start is to understand current trends of overall application consumption and bandwidth usage. You need to get a handle on historical data for which apps are being consumed, by whom, where in the network (for example, branches vs data centres or cloud), and by what type of networks (MPLS, Internet, and with or without VPN). This will give you a baseline from which to work and show you how scalable a solution you’re going to need in order to keep a close watch over how those applications get delivered to the business.
Fifth, and finally: What is comprehensive end-to-end application visibility worth to your organisation?
The answer to this question can be quite complicated. For example, even if you don’t want to pay for a robust, full-featured troubleshooting solution, you will still need to figure in the costs of your IT people’s time (and it’s never just one person, there’s usually a ‘war room’ full of people!) and effort to do the job manually. A complete ROI analysis also needs to include the costs and risks associated with slow or unavailable applications, which in many cases may directly impact service delivery or damage your organisation’s reputation. Matching the right solution to the right price to the right priorities will be key.
Of course, having asked those questions I am going to point out that Riverbed’s cutting edge SteelCentral and SteelHead, when used together, provide an integrated solution giving real-time application intelligence and network visibility – whilst at the same time accelerating application delivery and enhancing end-user experience. But that is all I will say. If you want to see a quick overview of how this can help you ‘detect, fix and relax’ on network problems have a look at this short video: