Introduction
We used pingtable.pl to look at the "Directivity (Alpha)" of connections from UNIMAS to other hosts in S.E. Asia.
We observed from UNIMAS:
->W MY A~0.45
->PH via JP A~0.22
->ID via HK A~0.17, traceroute via Telecom Malaysia
->TH via SG A~0.3, traceroute via Telecom Malaysia, SINGTEL
->BR via KL-HK A~0.07, traceroute via Telecom Malaysia and PCCW Global (HK)
->KH via HK-VN A~0.16
->VN, traceroute via Telecom Malaysia and PCCW Global (HK)
->SG A~0.23, traceroute via MYREN and TEIN3
Apart from Western Malaysia the routes looked very indirect (Alpha < 0.45)
The routing was initially estimated using the TEIN3 map (see below) and the PCCWGLOBAl map (for Brunei).
TEIN3 map for S.E. Asia |
PCCW Global map for S.E. Asia |
---|---|
|
|
see http://www.tein3.net/upload/pdf/1211_TEIN3_Topology_02.12.pdf |
see http://www.pccwglobal.com/images/stories/brochures/inf_map_lk_201105.pdf |
However according to the traceroutes only the route to Singapore actually uses TEIN3. This is also the only route to use MYREN. All the other routes use Telecom Malaysia. Instead of the TEIN3 map we therefore used traceroutes.
Looking at the routes using Telecom Malaysia (58.26.240.61), it can be seen that there is a lot of variability in the RTTs reported by this node. This bears further investigation, it is possible this node is congested with considerable queuing for several of the routes.
Country |
ms |
ms |
ms |
Avg |
Stdev |
---|---|---|---|---|---|
VN |
13 |
5 |
4 |
7 |
5 |
BN |
46 |
53 |
69 |
56 |
11 |
TH |
11 |
74 |
100 |
62 |
46 |
ID |
13 |
12 |
1 |
9 |
7 |
For more on the Indonesian links see here.