Applications delivered across the wide area network (WAN) are increasing in volume and complexity, and as a result organisations are shifting their approach to network performance management. The first concern for network engineers and operations managers is no longer just device availability, but also how well business-critical applications perform over the network. A review of the evolution of network management will help explain why this is happening and what it means.

Managing Network Availability
For most of the past two decades network engineers have focused on managing the network for availability. Today, however, most enterprise and service provider networks operate with 99.9 percent uptime or better. What drove the change? One reason is that technology has become more reliable in recent years. In addition, more and more mission-critical applications today are networked, and managers have realised that network downtime can be very expensive to the business. This high penalty has made it worthwhile for network managers to invest in redundant servers, hot-swappable components, and other hardware to guarantee uptime. 

Managing Network and Application Performance
While the reliability of networks has improved, performance issues have increased dramatically. The biggest influence driving this trend is the now common user expectation of a ubiquitous and instantaneous network. Other trends that have increased the volume and complexity of traffic and the need to monitor how well applications are delivered over the WAN include:

? Data centre consolidation: Enterprises and government agencies alike are migrating applications and data to central locations to save real estate, infrastructure, power and personnel costs, and improve manageability.

? Increased number of remote users: Branch offices and telecommuters are proliferating as companies grow, merge, and expand globally. Also, the exponential growth of e-commerce transactions continues unabated.

? The rise of voice and video traffic: Voice and video traffic are increasing rapidly and both rely heavily on the quality and consistency of network delivery.

? Legacy applications: "Chatty" applications designed to run over local area networks often do not perform well when deployed over the WAN.

? Software as a service: More organisations are choosing applications, such as Salesforce.com, that are delivered over the Internet as a service, in place of internally hosted, client-server applications.

? More complex applications including Service-Oriented Architectures (SOA): Distributed architectures and the use of Web Services as a means to develop reusable software for rapid application delivery and easier maintenance introduce network traffic between the various application tiers and infrastructure components. 

The convergence of increasing WAN use with improved device availability is leading network engineers to put performance first when it comes to managing their complex networks. (See Figure 1) By shifting the focus from fault management?which is largely under control?to performance-based management, network professionals can improve application service delivery and make themselves more relevant to the business units they serve.

Figure 1( To view this please register to DCS at www.datacentresols.com/register.php )
 

Over the past two decades, efforts by IT organisations and infrastructure vendors have reduced infrastructure downtime to a minimum. The most important trend in wide area networking today is the growing need to manage application performance.

 
The Case for a New Management Approach
A recent Yankee Group report?aptly titled "Performance is the New Mandate for Network Management"?refers to a study on enterprise application management. The study found enterprises report an average productivity drop of 14 percent when experiencing performance degradations in Oracle, SAP, PeopleSoft, Siebel, and custom .NET and J2EE applications. The same study found that almost 70 percent of network managers said they are now responsible for isolating the cause of application performance issues, meaning that their responsibilities now extend beyond the network infrastructure. Clearly, there is a case for performance-based network management.

A New Focus for Network Management: Performance First
The Performance First paradigm inverts the traditional, bottom-up device monitoring approach and begins with top-down visibility into overall performance of applications running over the network. In this approach, infrastructure availability and utilisation are no longer the sole gauges of network health. After all, it makes little sense to focus 100 percent of network management efforts on the 0.1 percent or less of the time there are hardware or software infrastructure failures. Fault management is necessary, but not sufficient.

Figure 2 ( To view this please register to DCS at www.datacentresols.com/register.php ) 

A Performance First approach begins by measuring end-user response times, then analysing traffic flows and device performance to optimise the network and troubleshoot problems.

Three Critical Components of Network Performance Management

The Performance First paradigm requires monitoring three key metrics (as illustrated in Figure 2):

? End-to-end Performance Monitoring: To track, measure, and analyse application performance for all user transactions from end to end for insight into the end user experience and the source of any latency issues.

? Traffic Analysis: To visualise and analyse the composition of network traffic on specific links. This yields the information needed to redirect or reprioritise application traffic, or add capacity.

? Device Performance Management: To poll network infrastructure components and isolate the source of problems such as a busy router or a server memory leak.

A Performance First management approach starts by understanding and baselining end-to-end performance: What is normal performance for an application or a group of users? How does "normal" change during busy and off-peak business cycles? Which applications and users are experiencing poor performance? Where is the increased latency occurring? Is it in the network, server or application itself? What impact did the new application have on other application response times? Did the infrastructure upgrade deliver the performance boost expected?

If the cause of latency is isolated to the network, traffic analysis enables network engineers to understand the composition of traffic on specific links where latency is higher than normal or expected. This yields the information needed to redirect or reprioritise application traffic, or add capacity.

If the source of latency is isolated to an infrastructure component?a busy router or a server memory leak, for instance?network managers need device performance management capabilities to poll the device in question and pinpoint the reason so that corrective action can be taken.

If latency cannot be attributed to the network or the server infrastructure and can be shown to be isolated in the application itself, the network team is armed with the proof that will eliminate the typical finger-pointing between IT infrastructure and application teams and expedite troubleshooting.

The Benefits of the Performance First Paradigm
The Performance First approach helps IT organisations optimise the delivery of critical application services by:

? Proving the performance of applications running over the network
? Delivering consistent application performance and measure it
? Mitigating the risks from planned changes and unexpected events
? Making more informed infrastructure investments
? Working collaboratively and more effectively
? Troubleshooting problems faster to reduce MTTR

Summary
IT organisations can no longer manage networks in isolation from the applications they support. Traditionally IT staff built their network management practices around infrastructure availability and fault management. Today most networks are available more than 99 percent of the time and increasing user expectations for fast, trouble-free networked applications requires a shift from a device-centric to a performance-centric focus.

ShareThis

Tags: Applications, Hosting & Colocation, Design & Facilities Management, Networks & Telco , Security, Servers, Storage Networking

Read more White Papers»

More DCS TV»

More Audio»

Read more Web Exclusives»

Read more Magazine Articles»

Read more Supplements »

Keep up to date with the latest industry products, services and technologies from the world's leading IT companies.

• Virtualization Success Depends on Network Automation by Infoblox

• The Infoblox VMware vCenter Orchestrator Plug-in by Infoblox

• Automated Storage Tiering on Infortrend’s ESVA Solution by Infortrend

• Why Cloud Backup: Top 10 Reasons by Autonomy

• How to Measure the ROI of Cloud Data Protection by Autonomy

Latest IT jobs from leading companies.

Click here for full listings»