Transcript TSACC Test Area

Performance Objectives
for performance testing
on CJK NGN testbed
Hyungsoo Kim
([email protected])
The 7th CJK NGN Working Group Meeting
October 10 ~ 11, 2006, Tokyo, Japan
Contents
1. Why?
2. Problem statements
3. Study plan
4. Further consideration
The 7th CJK NGN Working Group Meeting
2
1. Why?
 The purpose of performance testing (Scenario 6) is to evaluate a
service would be achieved the required performance level.
 The required performance level (objectives) should be identified.
 Some performance parameters have been discussed in Testbed
Group : delay, delay variation, loss, etc.
 ITU-T Recommendations could be referred.
i.e., Y.1541 (Network performance objectives for IP-based services)
Performance
Parameter
Nature of Network
Performance
Objective
Class 0
Class 1
Class 2
Class 3
Class 4
Class 5
Class 6
Class 7
IPTD
Upper bound on the
mean IPTD
100 ms
400 ms
100 ms
400 ms
1s
U
100 ms
400 ms
IPDV
Upper bound on the
1-10-3 quantile of
IPTD minus the
minimum IPTD
50 ms
50 ms
U
U
U
U
50 ms
50 ms
IPLR
Upper bound on the
packet loss
probability
1*10-3
1*10-3
1*10-3
1*10-3
1*10-3
U
1*10-5
1*10-5
IPER
Upper bound
U
1*10-6
1*10-6
The 7th CJK NGN Working Group Meeting
1*10-4
3
2. Problem statements
 Table of Y.1541 refers to Service Level Requirements, independent of actual
distance and number of links.
 At least 2 references are existing; TTC & TTA for VoIP service requirements
 End-to-end delay of Class A (TTC) : ~ 100 ms
 End-to-end delay (TTA) : ~ 150 ms
 If the path under test covers from Seoul, Korea to Tokyo, Japan, and the
performance testing result shows around 200 ms (90 ms of Japan side & 110
ms of Korea side), is it good quality ?
Y.1541 QoS class 0,2 & 6 < 200ms < TTC class C
 CJK NGN testbed will cover the distance from Japan to China through Korea.
(currently, from China to Korea)
 A note of Table 1 of Y.1541 introduces as follows :
 In some other circumstances (i.e., longer than 5000 km distance), the
IPTD objectives in Classes 0, 2 and 6 will not always be achievable.
The 7th CJK NGN Working Group Meeting
4
3. Study Plan
 Allocation rules of ITU-T Recommendation Y.1542 could be referred for
this purpose - Framework for achieving end-to-end IP performance
objectives
 Y.1542 recommends quite many approaches ;
- Static
- Pseudo-static
Which approach should be applied?
- Signaled
- Impairment accumulation
 The study considers various approaches toward achieving end-to-end (UNIUNI) IP network performance objectives.
① Analysis and choice of approaches; pros and cons
② Identification of distance for an end-to-end test path on CJK NGN testbed
③ Calculation of actual performance objectives for performance testing on
CJK NGN testbed
The 7th CJK NGN Working Group Meeting
5
4. Further considerations
 The real environment of CJK NGN testbed should be recognized.
 Relationship with national requirements in each countries’ policy should be
taken into account.
 Cooperation and coordination among 3 countries’ organizations (TTC, CCSA
and TTA) is needed.
 The international VoIP service requirements (i.e., service license) may
be considered.
 Performance allocation among providers on nation-wide may be studied.
The 7th CJK NGN Working Group Meeting
6