Q&A: Network Performance Management
Networks are more heavily burdened than ever with data, voice, and video. How can enterprises ensure the performance of business applications across the network?
Networks are more heavily burdened than ever with data, voice, and video. How can enterprises ensure the performance of business applications across the network? To get answers, we spoke with Patrick Ancipink, director of product marketing at NetQoS.
Enterprise Strategies: What is network performance management?
Patrick Ancipink: Network performance management (NPM) is the process of ensuring that applications are delivered across the network infrastructure as efficiently as possible and identifying where there is opportunity for improvement.
Traffic across the wide area network is growing in both volume and complexity due to trends such as software-as-a-service (SaaS), cloud computing, rapid adoption of voice and video, data center and server consolidation, and rising numbers of remote employees and branch offices. As a result, the performance of networked applications has taken on greater significance, especially for large, distributed organizations that depend on optimal application performance to conduct business.
How is NPM different from network fault management?
Network fault management focuses on the availability of systems, which, while important, is no longer adequate for ensuring that applications are performing well across the network. This approach focuses on devices by using fault management tools and processes to maximize network availability. Infrastructure reliability has improved to the point where 99.9 percent availability is not uncommon.
The NPM approach recognizes that between the limits of the network and application infrastructure being “up” or “down” there is a wide spectrum of performance variation. It is not uncommon for availability status indicators in the network operations center (NOC) to all be green -- meaning all devices are on and working -- while the help desk phones are ringing off the hook with users complaining about slow application response times. Traditional network fault management is important but not sufficient. Network performance management is better suited to help organizations address the complex performance challenges facing them today.
What is the business challenge that NPM is trying to solve?
NPM addresses the challenge of efficiently and cost-effectively delivering the applications organizations need to conduct business. Done well, NPM enables organizations to make informed infrastructure investments to support business demands, mitigate risk from both planned and unplanned changes, more quickly resolve problems, and deliver consistent, acceptable end-user response times.
What are the key metrics involved in measuring network and application performance?
One of the leading indicators in NPM is how well applications are performing for users. Organizations should measure response times for every critical application, establish what normal performance looks like, and receive alerts on where and when there are slowdowns. In addition, traffic analysis metrics reveal which applications are being used over the wide area network, who is using them, and when. This is crucial to understanding how the network is being used in multiple dimensions and quickly identifying the required steps to improve performance.
Deep-packet inspection capabilities with long-term storage enables past application transactions to be easily recounted. The ability to quickly rebuild complete Web pages with graphics and reconstruct e-mails, documents, and instant messages is important for forensics, compliance, and security auditing.
What are some of the best practices in NPM?
The adage “if you can’t measure it you can’t manage it” applies here, so capturing the right metrics is fundamental. An essential best practice is to focus on end-to-end application response times as the starting point for understanding how well the network is delivering services to end users. IT organizations can then drill down into other areas (such as traffic analysis and device performance management) as needed. Engaging in NPM best practices should enable you to easily answer the following five questions:
- Can you measure how well your mission-critical business services are being delivered? For instance, how is SAP performing over the network for your remote users? (you can no longer simply ask “Is SAP available?”)
- How much of the traffic on your network is business related and how much is not? As streaming applications such as real-time video become more prevalent, the ability to set quality of service (QoS) policies for which types of traffic get priority across your network is imperative. Ensuring the policies are working is essential.
- Can you regulate compliance initiatives with long-term network traffic storage and playback? As governments become more involved in technology regulation, IT organizations need to show proof of how the network is being used at any given time.
- Is your network prepared for a convergence of unified communications and data services? The addition of streaming application services such as VoIP and video can negatively impact the performance of data applications, and vice versa.
- Have you established service-level agreements (SLAs) around application delivery (not just availability)? Implementing application performance SLAs provides a measuring stick for how your end users receive application services and when performance degrades from normal. SLAs can also help you determine whether your external service providers are living up to their service guarantees.
Can you provide an example of how a customer would benefit from an NPM solution?
A large, global chemical manufacturer used network performance management tools and best practices to reduce network costs and improve application performance. Previously, this company used only gross bandwidth utilization reports, with no visibility into traffic composition across its 400 worldwide links. The company implemented NPM tools that showed which applications were being used on its global network, who was using them, and how much bandwidth those applications were consuming.
The organization was surprised to find that 50 percent of the bandwidth was being consumed by e-mail; 30 percent by Internet browsing; and 10 percent for print and other small jobs. This left less than 10 percent of the bandwidth for business-critical, time-sensitive applications. Even more alarming, the company found that it was purchasing twice the MPLS bandwidth that was required. To rectify this situation, the company began a project to classify all its global locations based on network demands and move them to the appropriate network infrastructure (MPLS or public Internet) based on the time sensitivity of applications.
How can customers justify investing in NPM solutions in a tough economy?
Regardless of the economic conditions, organizations still need to optimize application performance across the network and prevent unnecessary infrastructure spending. The complexity of delivering applications across today’s network infrastructures requires NPM processes and tools to monitor and manage performance rather than just availability. Best practice NPM should be leveraged to avoid and reduce costs in several areas:
- Optimizing bandwidth expenditures
- Eliminating overlapping and inefficient tools and equipment
- Reducing high severity mean time to repair (MTTR)
- Lowering the number of trouble tickets and support calls
- Reducing planning and decision cycles
- Improving end-user productivity (fewer “brownouts”)
The chemical manufacturer I mentioned was wasting millions of dollars on unnecessary bandwidth and did not realize it until the IT organization implemented NPM solutions.
How do NPM solutions fit with other infrastructure solutions, such as legacy network management platforms?
Traditional network management systems deliver useful capabilities but are generally not sufficient on their own due to the fault-based heritage. NPM solutions and legacy network management platforms should share data and integrate workflows to help IT groups simplify data collection, improve analysis, speed troubleshooting, and leverage their existing investments. Different stakeholders need different views and capabilities, but context and data should be shared as much as possible.
What kind of NPM solutions does NetQoS offer?
NetQoS provides network performance management software and services that improve application delivery across complex networks. We offer end-to-end application response time monitoring, network traffic analysis, device performance management, long-term packet capture and analysis, and VoIP and unified communications performance monitoring. We integrate this data in customized views in the NetQoS Performance Center management console to help organizations optimize application delivery, solve problems faster, mitigate the risks from change, and make the most efficient use of resources.