You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 14 Next »

Attendees

Anjum, Kashif, Johari, Nara, Abdullah, Bebo, Saqib, Les; Raja and Umar were invited.

MYREN in KL

Les and Anjum gave talks at MYREN on Dec 12th. Les' is available online, Anjum's is also available online.

Workshop in Kuching

The workshop brochure is here. The talks from Kashif and Les are linked to here.

UTM

Anjum visited University of Technology Malaysia (UTM) on the way back from UNIMAS to Pakistan. He met with Dr. Hanan.

During Anjum's last visit to UTM he had asked Dr. Hanan to deploy PingER monitoring node at UTM. Dr Hanan has now appointed Saqib (PhD student at UTM, email class786@gmail.com) to complete the deployment. Dr. Hanan will arrange for the public IP address. Dr. Hanan has also agreed to explore the possibility of having a PhD student work on the data auto-correlation part for link error detections once the monitoring node is set up.

Anjum also asked Saqib to attend the next skype meeting between SLAC and NUST on 10th of January (in Malaysia) so that he can get a better idea of the project.

Saqib/Dr. Hanan will let us know of the details once the node is setup. Hanan reports he is looking for a host and wil get an IP by Jan 4th, 2012.

UNIMAS monitoring host

There is a problem with the UNIMAS traceroute. It appears pinger.unimas.my is behind a NAT that is not passing the traceroute UDP probes. Johari has sent email to the network/system, admins, but does not have a response, He followed up with a reminder. The traceroute/ping server is important for the TULIP trilateration project. If there is no resolution from the network/system people then they will move the server to the DMZ.

Les has started a report on traceroutes from UNIMAS.

We need traceroutes from UNIMAS: to Peninsular Malaysia; to a host in the Philippines; and to Cambodia. Johari will measure these.

It is interesting that it appears UNIMAS does not use TEIN3 or MYREN apart from the route to Singapore. All the other routes Les measured go via the commercial ISP Telecom Malaysia.

Also for the Telecom Malaysia routes it appears there maybe some congestion at 58.26.240.61 since the RTTs vary a lot. This is not conclusive since routers give high priority (do it in hardware) to routing packets and lower priority to traceroute packets.

Telekom Malaysia, tested the 500 MBps line from 3rd to 9th of Dec. Hopefully we can gather some info once the data is collected from the PingER monitoring host. No major effect was seen (about a 10-14% improvement in RTT). It is not possible to turn off the traffic shaping. A performance  test would be to set upiperf at NUST and UNIMAS and measure the TCP performance memory to memory. Other measures can be made: using Speedtest (to a Speedtest site as close to UNIMAS as possible to eliminate long RTT challenges again from a browser at UNIMAS.) from a browser at UNIMAS; using an NDT server. 

Grants

Abdullah raised a question on grant applications. Anjum has sent 3 proposals with significant research components as examples. These need to be reviewed by Nara, Johari, and Abdullah to see how they may be appropriate for the Malaysian eScience program. We discussed potential research directions on the last day of the workshop. Notes from the meeting are available. The action items are:

  • We will pursue the Exploratory Research Grant for Initial data in order to support setting up three or more PingER monitoring hosts in Malaysia. Potential sites are UM, MYREN, UTM, and UNIMAS. This will the basis of a larger deployment in Malaysia. This in turn will provide end-to-end active current and archival monitoring data. This can be mined for research in:
    • understand the intra-Malaysia end-to-end Internet performance, identifying
      • bottlenecks,
      • poor performance and reasons
      • making recommendations for better performance
    • comparing Intra-Malaysian Internet performance with other regions
    • looking for correlations across multiple routes
    • Identifying anomalies

The infrastructure can also be used to provide landmarks for measuring RTT to target hosts for geolocation research.

Anjum shared proposals from NUST can be used as input to various proposals.

Johari will review the proposals and put together a draft.

UM

Abdullah has passed the information on PingER to colleagues. There are 2 people who are interested in pursuing. One will be at the MYREN workshop making a presentation. Abdullah will send email introducing the UM KL person. Is there an update?

NUST

From Arshad: Rector has agreed to support one person visit to SLAC. Raja has been nominated. He has sent a cv to Les and has filled out the DoE form. Les crafted the invitation and it has been sent to upper management for signature Dec 21, 2012.

PingER at SLAC

Les is working on a high level paper on PingER to be included in the IEEE Spectrum, February issue. He is also leading the preparation of the annual report to the International Committee for Future Accelerators (ICFA) Standing Committee on Inter-regional Connectivity (SCIC). Anjum is providing information on Pakistan under a case study. The report is due to Harvey Newman for ICFA on January 18th.

There is a draft case study of routing from UNIMAS to S. E. Asia, and in particular Malaysia .

TULIP

Raja got married 25-26 December, so there is nothing to report since the last meeting.

Future meeting time - Les

Next meeting Wednesday 23rd January 2013 8:00pm Pacific Daylight Time, Thursday 24th January 2013 9:00pm Pakistan time, Thursday 24th January 2013 12:00 noon Malaysian time.

  • No labels