NetQoS Takes "Performance First" Approach to Net Management
In our first four tutorials in the "Working Smarter" series, we have considered the five Specific Management Functional Areas: fault, accounting, configuration, performance and security management; reviewed some of the classic network management architectures, such as IBM's NetView and the IETF's Simple Network Management Protocol (SNMP); considered some of the challenges that can occur at each of the seven layers of the Open Systems Interconnection (OSI) Reference Model; and reviewed some of the devices that we need in our toolbox to address these challenges.
Now that we have outlined the network management challenges, it is time to look at solutions. This tutorial begins the second part of our analysis: considering the Windows-based solutions from one vendor at a time, beginning with those from NetQoS.
NetQoS, Inc., headquartered in Austin, Texas, is a privately held network performance management company whose software and services help large enterprises, government agencies and service providers--including half of the Fortune 100--improve the delivery of applications over wide area networks. NetQoS was founded in 1999, and has over 250 employees worldwide that support research and development centers in Austin and Raleigh, North Carolina, plus regional sales offices in London and Singapore. The company claims to have around 1,000 customers, including American Express, the American Heart Association, Barclays Global Investors, Boeing, Chevron, Deutsche Telekom, Dish Network, Hilton Hotels, Lockheed Martin, NASA, Siemens, and Walgreens.
The company was founded to provide better ways of monitoring and improving application delivery, in addition to developing solutions that could scale to support very large networking infrastructures. In contrast to the more traditional fault-based network management model, NetQoS takes a performance-first approach to managing network infrastructures, including systems that provide for end-to-end performance monitoring, traffic analysis, device performance management, IT cost accounting and retrospective network analysis. In addition, they have partnerships with other vendors, including Microsoft and Cisco Systems, to embed the NetQoS code and/or monitoring functions into these third-party products.
The NetQoS Performance Center product suite is a complete network performance monitoring and management solution with elements that are Windows-based, and are designed as appliance-based thin clients. By pre-loading its software onto off-the-shelf Dell or HP servers, NetQoS lowers its support costs and delivers enhanced customer service. Customer deployment is faster, with fewer installation and configuration issues, because NetQoS does not have to help the customer build and install hardware to run the software. The NetQoS solution supports LAN, WAN, MPLS and VoIP infrastructures, and can scale to the largest enterprise networks, with the ability to collect and store extremely large volumes of data without impacting product or network performance. The system includes capabilities for five key network management metrics: application response times, long-term packet capture and analysis, network traffic analysis, device performance analysis and unified communication Quality of Experience (QoE).
Application Response Times monitoring is achieved by a passive, server-side appliance that installs on a mirror or span port of a data center switch, passively monitoring every TCP/IP application. This function provides service level reporting down to the packet details, including WAN optimization monitoring and problem isolation through baselines and automated investigations.
Long-term Packet Capture and Analysis operates like the DVR connected to your television, allowing expert packet analysis with data stream reconstruction, forensic reporting, root-cause problem resolution, and detailed, multi-hop analysis. Up to 200 TB of network data can be stored for subsequent analysis.
Network Traffic Analysis gives you complete visibility into the protocols, hosts and conversations everywhere on the network by enabling NetFlow on key aggregation points, thus eliminating the need to deploy and manage network probes. (NetFlow is a protocol from Cisco Systems that collects IP traffic information). This function includes real-time enterprise-wide reporting, historical trending, and network-focused behavior analysis.
Device Performance Management includes SNMP-based device data, integrated with application response time and traffic composition. In includes service level agreement active response time tests, plus statistical projections for capacity planning.
Unified Communications Quality of Experience uses a passive appliance to collect metrics from the Cisco Unified Communications Manager, gateway, VoIP phones, and/or Microsoft Office Communications Server to illustrate the VoIP and video quality throughout the network. This application has the ability to watch critical calls in real time, and report Mean Opinion Score (MOS) and other key VoIP metrics.
The Performance Center includes a number of reporting options that tie all of the NetQoS capabilities together into a single interface for problem discovery down to root cause (see Figure 1). This figure illustrates the key network information that an executive might require, including transaction times, interface utilization, host activity and active protocols on the network.
Further details on the NetQoS architecture and products can be found at www.netqos.com. Our next tutorial will continue our examination of vendors' network management architectures.
Article © 2009 DigiNet Corporation, All Rights Reserved
Mark A. Miller, P.E. is President of DigiNet Corporation, a Denver-based consulting engineering firm. He is the author of many books on networking technologies, including Voice over IP Technologies, and Internet Technologies Handbook, both published by John Wiley & Sons.