ITU

Committed to connecting the world

Question 10/11

Question 10/11 – Service and networks benchmarking measurements
(Continuation of Questions 10/11 and 11/11)

Motivation
The issues of finding the optimum values of network performance and network productivity estimation approaches for services with required quality of service (QoS) is one of the important questions for the operators at the moment. The QoS/quality of experience (QoE) parameters and requirements are defined by ITU‑T SG 12.
 
The diversification of the system-network solutions, services and technologies cause some problems at the operator's networks. It depends on the performance estimation approach of the vendor's service platform which is not standardized and is absent at the moment.
 
At the moment, the common research part is QoS and QoE estimation approach which is based on the different estimation approaches from intrusive algorithm to the new approach directed to estimation and testing of the network performance values for the required QoS. This approach was titled – benchmarking. Benchmarking is based on the measurement and testing of parameters of network performance like latency, packet loss, setup delay, etc., which should help to save the required connection quality and network stability.
 
Question
Study items to be considered include, but are not limited to:
  • How model network architecture (determined in Recommendation ITU‑T Q.3900) have to be modernized for estimation of the optimum values of the network performance (NP) and services (benchmarking)?
  • What kind of services could be tested?
  • What kind of testing scenarios could be applied for estimation of the optimum values of the NP and services?
  • What type of traffic could be simulated for estimation of the optimum values of the NP and services? What is the set of traffic's type?
  • What kind of testing scenarios could be applied for estimation NP parameters for L2/L3 VPN services?
  • What kind of etalon of services and the profile of services could be used for estimation of NP parameters?
  •  What kind of the QoS/NP/QoE parameters included to service level agreements (SLA) and responsible for the QoE of services could be monitored?
  • What kind of testing scenarios could be used for SLA monitoring?
  • How test cases which are determined in the existing Recommendation ITU‑T Q.39xx-series have to be modernized?
  • What type of testing scenarios could be applied for interactive services?
  • What type of testing scenarios could be applied for stream services?
Tasks
Tasks include, but are not limited to:
  • determine the model network architecture for estimation of the optimum values of the NP and services;
  • determine the types of services, which could be tested;
  • develop the test specifications for estimation of the optimum values NP parameters for provided telecommunications services;
  • determine the set of the simulated traffic's type for estimation of the optimum values of the NP and services;
  • develop the program and test specifications of estimation of the optimum values NP parameters for L2/L3 VPN services;
  • develop profiles of the services;
  • determine types of the SLA which could be tested;
  • develop the SLA testing/monitoring approach for estimation of the services QoE and optimum values NP parameters for supporting required level of the QoE parameters;
  • determine the set of the service performance test cases;
  • develop the program and test specification of the interactive services;
  • develop the program and test specification of the stream services.
An up-to-date status of work under this Question is contained in the SG 11 work programme
 
Relationships
Recommendations
  • Q, Y, H, I, M, F, P and G-series
Questions
  • 1/11, 2/11, 3/11, 4/11, 7/11 and 12/11
Study groups
  • ITU‑T SG 13 on NGN architecture
  • ITU‑T SG 16 on multimedia services and applications
  • ITU‑T SG 12 on QoS/QoE, especially Questions 1/12, 2/12, 11/12, 13/12 and 17/12
Standardization bodies
  • IETF, ETSI, IEEE
  • Regional SDOs and other bodies dealing with NGN
 

 

​​​​​​​