Network Troubleshooting in Today’s Distributed Data Center
by September 22, 2011

Filed under: Network Performance Management, Performance Management Tech

Today’s distributed data centers, with their layers upon layers of virtual and physical devices and systems, are dauntingly complex and getting more so every day. The accelerating adoption of IP-based services like cloud computing, virtual desktop infrastructure and unified communications is further compounding this complexity, while also massively increasing the scope of the network.

All this growth and change confounds troubleshooting, especially where network performance and application availability are concerned. It is well known that when network performance drops even slightly below tolerance thresholds, IP-based services quickly begin to falter and collapse. Worse yet, it is becoming more apparent based on experience across the IT industry that poor network performance has the potential to cause outages to “cascade” through the data center, with unpredictable and unpleasant results.

Distributed Data CenterIn today’s intricate environments, elusive network and application problems are capable of causing major havoc to business operations. Even IT leaders like Microsoft, Google and Amazon have had their public cloud infrastructure knocked offline by complex, network-related issues in recent months.

Charged with meeting application service levels in the face of accelerating change and burgeoning throughput demands, today’s network administrators need better visibility into network performance – across LANs, WANs, all relevant flavors of virtualized/cloud infrastructure and even the internet. The ability to find the root cause of network problems before they topple the whole IT edifice has never been more critical.

Likewise, the ability to monitor performance in real-time across the distributed network is essential to proactively addressing problems before they impact users. In order to prevent bottlenecks from forming, IT staff needs visibility, alerts and reporting on trends affecting key performance metrics like capacity, throughput, latency, packet loss and jitter. Of course, given the realities of frozen or shrinking IT budgets, new technology needs to offer rapid time-to-value and reduce, not increase, administrative complexity.

Traditional network and application performance management solutions can’t offer adequate visibility into today’s dynamic, distributed IP networks. It remains especially difficult to gain insight into how network dependent applications are performing from the standpoint of distributed users.

What network engineers need is a single interface that allows them to continuously monitor all the key metrics that impact application performance across the entire distributed network, including elements serving virtual infrastructure, and even over public and third-party networks, from any location. These capabilities also enable IT to accurately assess the impact of new services on network performance, before rolling them out.

AppNeta’s PathView Cloud network performance management solutions provide unmatched insight into key network performance metrics like latency, jitter and packet loss across your entire distributed data center infrastructure. PathView Cloud radically simplifies the task of managing, troubleshooting and optimizing distributed network performance. Delivered as a hosted service, PathView Cloud is affordable, easy to deploy and offers near instant time-to-value.