Support Systems: Making VoIP Better

New Monitoring, Testing Tools Help Assure the Quality of IP Voice

Tollgrade’s fully integrated Cheetah IP/HFC Service Assurance Solution, including the CheetahIP element management system and hardware probes, allows operators to remotely and proactively test and monitor VoIP and VoD performance while maintaining the integrity of other critical applications.

VoIP monitoring and quality assurance are becoming hot areas as IP-based voice services become more mainstream. A significant number of vendors now offer tools specifically to help service providers and businesses to monitor, analyze and test their IP-based voice networks and services.

According to a new study by Frost & Sullivan, the test equipment market for VoIP presents one of the best opportunities for communication test vendors in recent years. The industry generated revenue totaling $133.2 million in 2003, according to the study, and total revenue is expected to reach $606.9 million by 2010. While the bulk of testing requirements now focus on the R&D applications of VoIP, the study says, the growing installed base of networks is expected to increase demand for monitoring and maintenance solutions.

VoIP test and monitoring vendors say that monitoring and troubleshooting VoIP services present unique challenges because many voice calls still traverse both IP and PSTN networks; different companies might provide the voice service and the broadband pipe over which it travels; calls can take winding paths on IP-based networks; and various protocols and vendor equipment types may be involved.

“With VoIP, problems can occur in various places in the network, so you need smart technology that’s embedded in the network to collect information when and where it happens,” says Alan Clark, president and CEO at Telchemy Inc., whose VoIP monitoring and management software can ride on network infrastructure like softswitches or on test or monitoring solutions.

Telchemy sells VQmon, which does call quality analysis; monitoring of the voice stream, looking at delay and other parameters; measurement/ ranking of call quality; and shows where degradation is, if it’s present. The company’s VQexpert can recognize LAN and access line congestion. And a technology called VQmon/EP from Telchemy addresses enterprise IP telephony, broadband VoIP and 3G applications.

The company’s software is currently in use by at least 15 infrastructure and support system vendors, including Acterna LLC, Brix Networks and Nortel Networks, which is using Telchemy software in its enterprise IP telephony gear, says Clark. He adds that one of the RBOCs is now using Telchemy technology and says that Telchemy also is working with a session border controller vendor to enable service providers to monitor the quality of calls handed off to their session border controllers from other service providers.

Quintum Technologies Inc. also is a believer in embedding monitoring capabilities in VoIP infrastructure equipment. The company, which launched its Tenor multipath switches in December of 1999, recently added monitoring capabilities to its Tenor devices. The devices can now do real-time monitoring of the status of alarms on Tenor equipment; monitoring of call events to see where calls are going and originating; and monitoring of call service records.

Jamie Warter, vice president of marketing and business development at Brix Networks, which sells VoIP performance management and service assurance solutions, says that the company’s first two customers were iBasis and Level 3 Communications Inc., who used Brix tools for IP voice transport and toll-bypass applications. But new Vonage-style VoIP entrants that are using local broadband connections to deliver packet voice services can also use Brix tools to prequalify VoIP services on residential links.

In this case, the Brix software looks at both the quality of the user’s broadband connection and the quality of the call running across that connection.

“That’s where we bring value,” says Warter, “so the service provider knows if it’s their infrastructure or the transport provider’s problem.”

At the VON show this spring, Brix demonstrated a self-service VoIP portal that lets users independently measure the quality of their VoIP connections. TestYourVoIP.com allows visitors to execute a JAVA applet that initiates a test phone call based on SIP. Brix Verifiers emulate multiline phones that answer the test calls and measure their quality for such metrics as signaling quality and call quality/clarity. The service, according to Brix, is aimed to help prospective VoIP consumers evaluate their expected call quality before signing on for service, to give customers leverage if their existing VoIP provider is not living up to expectations, and to help service providers attract and keep customers by demonstrating their high-quality service.

Warter explains that it was just a demo, but that service providers might want to employ Brix technology to create their own portals offering this user-testing capability.

Switch Management Corp. also offers a Webbased VoIP monitoring tool, which it launched in April. VoIP Watchdog is a Web-based quality of service monitor and alarm manager that alerts network managers when route quality (VoIP or TDM) drops below acceptable levels. A hosted application that is available as a stand-alone service or offered in conjunction with Switch Management’s WebCDR billing service, VoIP Watchdog continually scans a network’s most recent call detail records for alarm conditions like low answer seizure ratio, low average length of call, or too many consecutive incomplete calls on a route. Scans are performed every 15 minutes, and are immediately rated and analyzed. An e-mail alert is generated whenever alarm conditions are detected, and can be directed to cell phones or alphanumeric pager services that provide an e-mail gateway. Global Direct Telecom has been beta testing the service since March 1.

Tollgrade Communications Inc., whose CheetahIP product for cable network operators employs Brix technology, is also addressing VoIP service quality. Greg Quiggle, executive vice president of marketing at Tollgrade, says several cable TV companies are looking at its technology to troubleshoot VoIP problems.

Today, when a customer using VoIP over a cable modem reports a problem, there is no way of knowing whether the problem is with the plant, the power supply, the DOCSIS connection or the voice path, says Quiggle. But CheetahIP can look at all those aspects of a VoIP service so the service provider can quickly detect the source of the problem, he says.

CheetahIP already is used on a broad scale by major cable companies including Cablevision, Comcast, Cox and Time Warner, says Quiggle, and several MSOs are now testing it for VoIP applications. Tollgrade’s product traditionally had focused on the physical layer, doing status monitoring of hybrid fiber/coax plant nodes and power, but the Brix partnership has enabled Tollgrade to pair that capability with the ability to look at the quality of the DOCSIS connection to do VoIP performance monitoring, he says.

Network test giant Agilent also is a reseller of the Brix products and has integrated Brix technology into its NgN product, which Cbeyond Communications is using for network-wide VoIP quality monitoring.

Donna Bastien, Agilent’s worldwide OSS marketing communications manager, says scalability and performance have become key issues for VoIP as it moves into “prime time” with companies like AT&T offering residential local and long-distance IP telephony services. Jim Banks, VoIP business development manager at Agilent, notes the vendor recently announced the ability to monitor up to 1.5 million busy hour call attempts at one softswitch location. “That’s an indicator it’s moving to prime time,” he says.

The latest version of NgN Analysis System — which offers end-to-end monitoring over IP and PSTN voice networks — also can monitor multiple softswitches rather than just a single softswitch in a test environment, adds Bastien, explaining this is a requirement in real-world VoIP networks.

Agilent’s Analysis Servers are collocated with softswitches in a central office. They pick up all the signaling information from the softswitches and send it to the endpoints, feature servers and media servers. The NgN AS then stores that data, correlates it and sends it out to a central qualityof- service (QoS) management server in the data center that does service and management performance in real time. That allows the product to deliver real-time indications to the service provider if an SLA is out of compliance, for example. Agilent products also can look at media streams to monitor the quality of service delivered, says Banks, noting that Brix active test technology that is integrated into the Agilent products helps enable the QoS management capability.

Banks adds that today’s VoIP monitoring and troubleshooting tools give network operators a good idea about what endpoints are affected by service degradation, but they don’t provide much information about the virtual circuit. “The IP environment is very dynamic, you don’t know which routers you’re going to from one minute to another,” he says. An understanding of the network topology a service is using and the ability to correlate that network topology information with signaling and voice quality information could allow network operators to find causes of problems in a more efficient manner, he says. That eventually will be possible through MPLS, which can provide information on which paths in the network are used for what traffic, Banks says.

While more focused on the enterprise market, a company called Qovia also has products that monitor VoIP networks continuously in real time, looking at such parameters as packet loss, latency, jitter and even the capacity of the server providing music hold, the T1 and the power.

Acterna, meanwhile, recently unveiled the PVA-1000, which performs detailed analysis and troubleshooting of VoIP telephone calls, providing quick identification of VoIP transport-related issues by graphing jitter and packet loss. It plays back the audio of the original telephone call using jitter buffer emulation and WAV files, and also can provide full VoIP protocol decodes.

And earlier this year Spirent Communications introduced Abacus 5000 IP Telephony Test Migration System, which can emulate a call agent (media gateway controller) and signaling gateway, as well as generate real voice and data traffic and real-time transport protocol (RTP) packets.

But the solutions mentioned here are just a small fraction of a multitude of VoIP quality assurance products available from a variety of vendors today.

“VoIP is not as a simple as traditional data services. Dropped calls, latency and voice quality all come into play when mixing voice and data traffic on the same network, which today is usually some variation of an IP and PSTN network,” says Zeus Kerravala, a vice president at The Yankee Group. “Integrated testing is the only way to adequately verify the migration from circuit to packet to avoid performance, interoperability and voice quality pitfalls.”

Share this article: Email, Slashdot, Digg, Del.icio.us, Yahoo!MyWeb, Windows Live Favorites, Furl
Add this article feed to: RSS, My Yahoo, Newsgator, Bloglines