APAN 2003 Meeting Broadband Communication Lab. One Way Delay Measurement in KOREN MPLS Networks Sok-Hyun Jung [email protected] Broadband Telecommunication.

APAN 2003 Meeting

Broadband Communicati

on Lab.

One Way Delay Measurement in KOREN MPLS Networks

Sok-Hyun Jung [email protected]

Broadband Telecommunication Lab.Dep. of Electronics Engineering, Korea University

2 APAN 2003 MeetingBroadband Communication Lab.

Experiment Description

Objectives of ExperimentTo measure End-to-End One way Delay through various Transit networks such as Conventional IP and MPLS networksTo evaluate MPLS performance

Related projectWe had a grant of foundation of Korean National Computerization Agency Experiment organization

Korea University (KU), Chonnam University (CHNU)

3 APAN 2003 MeetingBroadband Communication Lab.

Overall Experiment Environment

One way delay measurement

GPS satellites

Transit Networks KU

KU connects with CHNU through various transit networks Such as Conventional IP and KOREN MPLS networksIn case of KOREN MPLS networks, LSP is set up between KU and CHNU

KOREN MPLS network

Conventional

IP network

Korea Univ.(KU)

ChonnamUniv. (CHN

U)Label Switched Path (LSP)

4 APAN 2003 MeetingBroadband Communication Lab.

KOREN networks

Korea Univ.(KU)

ChonnamUniv. (CHN

U)

Constitutions of KOREN ATM switches added MPLS function

LER & LSR

Functions of KORENCurrently, provides only basic functions such as LSP setup using the LDP (Label Distribution Protocol) and so on. In the future, provides TE, VPN, QoS and so on.

5 APAN 2003 MeetingBroadband Communication Lab.

Transit Network 1 – Conventional IP network

One way delay measurement

ConventionalIP network

CHNU KU

168.131.161.30 203.255.252.65

This network is compared with other transit networks such as Packet based MPLS and ATM based MPLS

6 APAN 2003 MeetingBroadband Communication Lab.

Transit Network 2 - Packet based MPLS

CISCO7204

Packet based MPLS network

CHNU KU

CISCO 7505

GSR RouterDaejon

GSR RouterSeoul

GSR RouterKwanju

Swap Table

X Y

Swap Table

Y Z

Swap Table

Z X

MPLS runs over the Ethernet as link layer protocolLSP is set up between CHNU and KU

168.131.161.30 203.255.252.65

Label Switched Path (LSP)

7 APAN 2003 MeetingBroadband Communication Lab.

Transit Network 3 - ATM based MPLS

CHNU KU

ATM based MPLS network

CISCO7204

CISCO7505(Nortel 15000 )

Seoul(Nortel 15000 )

Daejon(Nortel 15000)

Kwangju

(Fore ASX200BX)

(RealCell)

Allocating New VP

Unlike Packet based MPLS, ATM based MPLS makes use of ATM as link layer protocolNew VP is allocated from CHNU to KU and similar to LSP

168.131.161.30 203.255.252.65

8 APAN 2003 MeetingBroadband Communication Lab.

How to measure ?

1. To Synchronize between two end systems using GPS 2. Host1 sends one way traffics to host2 through various Transit networks 3. Host2 receives traffics, and measures delay4. To compare measured results for each Transit networks

GPS satellites

Transit NetworksHost1

Sender Receiver

AMS: Active Measurement System

AMS

CHNU KU

Host2

AMS

9 APAN 2003 MeetingBroadband Communication Lab.

Result 1 : Conventional IP network

Table1: One-way Delay Measurement for conventional IP at KU

Table2: Ping test for conventional IP at KU

One way delay is about 10 msec

10 APAN 2003 MeetingBroadband Communication Lab.

Result 2 : packet based MPLS

Packet based MPLS networkCHNU KU

ATM ATM ATM ATM ATM ATM

168.131.161.30 203.255.252.65

Table3: One-way Delay measurement for packet based MPLS at KU

Table4: Ping test for packet based MPLS at KU

Host1 Host2

CISCO Router(LER)

CISCO Router(LSR)

CISCO Router(LSR)

CISCO Router(LER)

One way delay is about 5 msecIt is smaller than that of the conventional IP network

11 APAN 2003 MeetingBroadband Communication Lab.

Result 3 : ATM based MPLS

ATM based MPLS networkHost1 Host2

CHNU KU

ATM ATM ATM ATM ATM

CISCO Router(LER)

CISCO Router(LER)

ATM

168.131.161.30 203.255.252.65

Table5: One-way Delay Measurement for ATM based MPLS at KU

Table6: ping test for ATM based MPLS at KU

One way delay is about 3.6 msecIt is smaller than that of the Packet based MPLS

12 APAN 2003 MeetingBroadband Communication Lab.

Conclusion

Delay in ATM based MPLS is the smallest value ATM based MPLS can reduce one way delay, as compared the Packet based MPLS and Conventional IP network

Case 1

Case 2

Case 3

Conventional IP

Packet based MPLS

ATM based MPLS

10.995 msec

4.995 msec

3.578 msec

Types of Transit Networks E2E One Way Delay Bad

Good

13 APAN 2003 MeetingBroadband Communication Lab.

Thank you for your attention !!

Thank you for your attention !!