A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 0-9

Pour ajouter des favoris à une liste, vous devez être connecté à votre compte.

Me connecter Créer un compte

Publié le 19 mai 2015

Choosing the right TWAMP-based SLA Monitoring Solution

Are you an operator who has opted for the Internet protocol (IP) as your preferred transport layer? Are you currently using ICMP Ping/TraceRoute or UDP Echo tools to measure your IP network? Are you looking to learn more about the Two-Way Active Measurement Protocol (TWAMP) and how it can improve your SLA monitoring?

Today, IP network operators are looking for more precise, accurate and relevant tools to ensure carrier-grade performance. The Internet Engineering Task Force (IETF)’s working group on IP Performance and Metrics developed RFC-5357, the TWAMP. The TWAMP protocol is a standards-based and highly effective performance monitoring process that that expands upon the One-Way Active Measurement Protocol (OWAMP) specification defined in RFC-4656 with the addition of the performance measurement of round-trip and two-way metrics for IP based networks. TWAMP is a vendor-agnostic method for accurately measuring unidirectional and round-trip performance between two TWAMP-supported endpoints.

But just what is the TWAMP and what features should you look for in a SLA monitoring solution? Let’s take a closer look.

The TWAMP protocol includes the following components: test endpoints, two inter-related protocols, namely TWAMPcontrol and TWAMP-test, and and Modes (Light or Full).

The overall architecture of TWAMP is comprised of endpoints with specific roles that simplify the measurement of IP network performance: the session-sender node, which is responsible for generating test traffic and processing the returned traffic in order to measure performance, and the reflector node processes, which returns test traffic received from the sender node. Several solutions on the market exist; however, many can only respond to TWAMP traffic—and not generate it. Only a few feature both sender and responder capabilities—which is ideal if you are looking for maximum flexibility to troubleshoot your network.

The TWAMP process is defined in two different modes: Full and Light mode. The TWAMP Full mode test is designed to work in a client-server relationship, where the control of the test may be managed by separate devices from the devices that will be sending and receiving the test traffic. TWAMP Light mode is designed to help implement the TWAMP standard across entities that act as active responders to TWAMP controllers within the network, thereby enabling the measurement of two-way IP performance from anywhere within that network. The difference between the two is that during a Full-mode TWAMP session, the session is established between the sender and responder through a control channel within a control session that negotiates communication between devices. In a Light-mode TWAMP session, no negotiation occurs between the endpoint and the client. While each mode has its advantages, both measure critical KPI’s, including loss, latency, jitter, duplicates, out of order scenarios, and more. Again, when selecting a SLA monitoring solution, be sure that the one you opt for has both for optimal monitoring flexibility.

There are other considerations that you should take into account when selecting a SLA monitoring solution for your IP network. Does the product:

  • Enable testing of both periodic and continuous sender traffic? Periodic TWAMP sender tests offer higher scalability of tests as multiple tests can be performed simultaneously while consuming less aggregate bandwidth as traffic is spread across measurement periods. Continuous TWAMP sender tests, on the other hand, enables a more granular and uniform measurement of performance as all measurement seconds will be filled with a steady stream of test packets.
  • Perform reflector tests? This test acts as a service that listens for data from the TWAMP sender traffic; it can process a high-rate of packets and support multiple senders simultaneously for increased testing efficiency
  • Trigger TWAMP-encapsulated activation tests? These tests measure the maximum capabilities of a service or a circuit by leveraging RFC-2544 and Y.1564 tests. The benefits of these tests is to take advantage of far-end responder capabilities to eliminate the need for service truck rolls.

To read more about TWAMP, specific use-case scenarios and what TWAMP-based SLA solutions are available for more accurate and reliable monitoring of IP networks, read our SLA Monitoring Using the Two-Way Active Measurement Protocol (TWAMP) app note!