Versions Compared

Old Version 9

changes.mady.by.user Les Cottrell

Saved on

compared with

New Version 10

changes.mady.by.user Zafar Gilani

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • The draft letters of invitation for Sadia and Amber have been sen, however they had the old dates, we are working on a new pair. They have the other docs (DS2019 etc.) Current plan is for arrival April 1st. 
  • Sadia, Amber and Bilal have their SLAC computer and Wiki accounts. Amber had a problem with her Windows password. The new confluence went into operation. It is now working for Amber.
  • Zafar received his DS-2019 on Dec 10 (Fri). Applied for visa on Dec 13 (Mon). Interview date was Jan 3, it has been delayed to Jan 24 Jan. Visa accepted now await clearance. Expect ~1 weeks.  He has not recieved his visa yet. Hope to start mid march. 

Pakistani case study - Zafar, Anjum

  • Pakistani case study is done for the moment. Since I will not be visiting Pakistan in the near future, I have sent an email to Sohail Naqvi of HEC telling him of our case study of Pakistan.  Now we need to work with Arshad to decide how to present the results to PERN/HEC.  Anjum will talk to Arshad today and after today's meeting then will present the case study to the PERN folks to try and understand the problems. PERN are interested in losses, reachability, and RTT. They are also interested in perfSONAR throughput measurements. For NUST this was waiting on the 10Mbps link. This had to go thru NUST who rejected the idea. It is too expensive. There is another link for the research purposes which will be split into 2, part for perfSONAR. PERN will deploy perfSONAR at HEC/Quetta. Someone is working on this. The university is close by HEC/Quetta. Hope in 4 weeks to have PingER monitoring node in 4 universities. More of a research project than a deployment project. 
  • HEC wants the ability to view just the Pakistan results. Want a view of whatever we are recording. They can use the map to see the topology. Faisal will look at a new user interface with a Pakistan bias.
  •  Zafar has started taking and archiving traceroutes between Pakistani monitoring hosts on a daily bases. WE  We need a user interface to select and view these traceroutes. 
  • The UETTAXILA to NUST report has been updated. However Zafar stopped the pings. he He will re-start so we can get measuremens made when the university is in a stable state. During daytime until 3:30pm when the classes are on, internet access is blocked to the hostels. After 3:30pm it is enabled thus RTT increases. Amber will update her description of how this affects the RTTs.
  • Sadia is working on prm.pl to add alpha and MOS. I have fixed the creatpage.pl problem. An example of the output from the new prm.pl n_prm.pl is at: 

...

PingER

  • We are working on getting pinger2.pl working at Sarawak. The pinger.xml page needs to be fixed
  • PingER archive site: Ghulam and Farhan  have  created have created two ERDs. One is a simple design (relatively inefficientthree tables, normalized to level 1) and another is a complex design (relatively efficient though it might introduce five tables, normalized to level 3, introduces some redundancy) with multiple normalized tables that can be joined. They have created MySQL databases following both ERDs and have done some stress testing. The simple design is faster to load, the more complex design is faster to insert data into and load data from. They will probably go with the more complex design. Umar went through the URDs, and requested a document on what they have done. The schemas have been shared but there is little justification given for the designs. They sent the analysis to Umar for the 2 schemas. The simple one with 3 tables is more efficient time and spacewise. The difference is quite large. They are trying to copy the files from the flat files to the the database. We will need to compare space used by the database compared to the flat files. they will need to modify pinger2, pingtable, possibly the analyze scripts
    • Simple design takes 15.4 seconds to write data from 15 files and 12 seconds to read. Complex design takes 18 minutes and 57 seconds to write data from 15 files and 13 seconds to read.
    • Simple design is also space efficient. It takes 27 MB of space as compared to 30 MB taken by complex design for writing data from 15 files.
    • As suggested by Umar, we should run more analysis before finalizing this.
  • Faisal is working on a home page for PingER.
  • PingER Map
    • Please add nodename and fullname.  (done)
    • Test Maximum Minimum values (done)
    • Add The ability to choose the line thickness (done)
    • Remove the loading bar (done)

...

  1. Faisal is working with Yee.## Datagrid module to display tests from all the result set hosts for a selected ## Ability for datagrid module to share data and calls with map.
  2. Documentation available here https://confluence.slac.stanford.edu/display/IEPM/Google+Maps+for+PerfSONAR|../../../../../../../../../display/IEPM/Google+Maps+for+PerfSONAR|||\
  3. Zafar - deploy deployed PerfSONAR nodes at SEECS. Second step is to find whether the same can be done at HEC on PERN's network. It is working, congratulations! Is it taking any data?. It was working and gathering data. However the Network Operations Center at SEECS had to shut it down since it was flooding the network and disrupting normal traffic (education/research). Sent an application to HQ NUST to approve 10 Mbps bandwidth link. The proposal was rejected. Now waiting to utilize 2 Mbps from a 10 Mbps research network link.

Possible projects

Wiki Markup
See \[https://confluence.slac.stanford.edu/display/IEPM/Future+Projects\]. Zafar will talk to the students about these projects.

...

The next meeting in Wednesday February 23rd 2011 for people in US and Thursday February 24th January for people in Pakistan.