Mu Dynamics Ensures Operators Remove Costly Problems, Limit IP Service Downtime

By: Richard “Zippy” Grigonis

Mu Dynamics (www.mudynamics.com) helps both network operators and product suppliers build and operate networks to ensure their IP services have minimal downtime. The new generation of IP services such as VoIP, IPTV (News - Alert) and IMS rollouts must be robust, available and secure to minimize customer churn. To that end, Mu Dynamics offers the Mu-4000 Analyzer, an ingenious, highly flexible system that enables proactive service assurance.

The Mu-4000 automates the discovery, documentation and expedites the remediation of any reliability, availability or security weaknesses. For example, Mu methodically generates millions of variations of service level traffic to analyze the impact on IP-based networks, identifying any downtime-causing weaknesses in such next-gen services as VoIP, IPTV and IMS-based applications.

David Kresse, CEO of Mu Dynamics, says, “We ensure operator reliability, availability and security of IP services and applications. We know IP is here to stay — there are many associated benefits. Cost advantages exist for running applications and services over a shared infrastructure, as opposed to ‘siloed’ infrastructures. In addition, the ability/flexibility to roll out new applications and services quickly is important as a competitive advantage.”

“However, IP has inherent challenges,” says Kresse. “IP infrastructures are inherently complex and more fragile than the last generation of infrastructure. As a result, businesses need to think differently about their approach to the reliability, availability and security of IP applications and services. Mu offers operators and their vendors a new solution to uncover weaknesses that compromise the quality of their service or application before deployment. Essentially we’re running lab-contained but real-life network service traffic. A customer can simulate what is going to happen once a service is deployed and proactively identify issues that cause service degradation or disruption. Given the costs associated with service disruption — customer churn, high volume of support calls, violated SLAs — the ability to find issues before they impact customers is of immense value. The ROI around deploying our solution is huge.”

“We call our system a service analyzer,” explains Kresse. “It consists of three different components: First of all, we can generate different kinds of simulated network traffic. For example, we can generate millions of variations of traffic around specific protocols used by services or applications. We also run simulations of known vulnerabilities as well as Denial-of-Service traffic profiles. Some customers use all three of these modules, others just use what they need. Although you’re working in a test environment, running our analyzer is like experiencing these situations in a live environment.”

“Our analyzer’s second major component comprises service-level monitors that carefully observe the services supported by the systems or device under test,” says Kresse. “Basically, as the Mu system generates application traffic, monitors automatically correlate specific variations in that traffic that adversely compromise services.”

“The third major component is a correlation engine enabling our product to precisely pinpoint what specific aspect or variation of application traffic is at the root of service degradation or downtime,” says Kresse. “It allows our carrier and cable operator customers — many of them are among the largest in the world today — to precisely identify the location of issues and to work with their vendors to remediate them before the services are deployed in their live network.”

Kresse elaborates: “As Mu spoke to its 100+ customer deployments about how they deploy our equipment, the key insight we’ve gleaned from a business perspective is that, when you look at IP environments, a change has occurred. Before IP infrastructures became commonplace, customers had relatively homogenous infrastructures from a vendor standpoint — a few vendors per environment — and a very controlled number of dependencies and interactions occurring between different components in support of a given service — generally a dedicated service. In that era, functional and load testing effectively dealt with the limited problem scope. Today, operators deploy IP services within a highly heterogeneous infrastructure, with lots of different components interacting with one another to support multiple services. The idea that ensuring valid inputs result in valid outputs and everything is going to work well is only a piece of the puzzle.”

Frequently in today’s increasingly complex networks unexpected inputs or even just slightly malformed packets emanate from one application or product. The question becomes: How will other networked products within the overall application or content delivery network respond? Will they elegantly handle unexpected traffic or simply end up ‘choking’, possibly exposing potential product weakness like a CPU spike, buffer overflow or unacceptable product latency when exposed to such unexpected input? Mu methodically identifies these weaknesses before the live deployment of applications and services to save operators from experiencing costly downtime, customer churn or expensive brand damage.

In all, Mu Dynamics appears to have mastered an important new area of network and services analysis. Operators have a new way to ensure their services deliver as promised and to hold their vendors accountable. IT

» Internet Telephony Magazine Table of Contents