Time & date
Meeting time Wednesday February 5th 2014 8:00pm Pacific Standard Time, Thursday February 6th 2014 9:00am Pakistan time, Thursday February 6th, 2014 noon Malaysian time, Thursday 6th February, 2014 02:00am Rio Standard Time.
Attendees
Invitees:
Anjum, Hassaan Khaliq, Kashif, Raja, Johari, Nara, Abdullah, Badrul, Ridzuan, Hanan, Saqib, Adib, Les, Renan, Bebo
+ Confirmed attendance
- Responded but Unable to attend:
Actual attendees:
Administration
- Johari proposes that UTM host a PingER workshop in 2014, maybe associated with another conference. Anjum will be in Malaysia then. Saqib will meet Hanan Jan 9th and discuss this. After the meeting Johari sent email to the Les, Hanan, Saqib, Badrul, and Adib with the issues and suggesting a skype meeting
UTM volunteered to host the workshop last year. Due to the lack of funding from UTM, UM hosted it. Hanan believes the constraint is the budget. After knowing how can we fund the workshop, then we can decide who will the host the workshop.
Adib suggested a summer school that is aimed at graduate students and researchers from different disciplines where everyone needs to be present for team presentations based on case studies/assignments given to each team. Each team will be mentored by one professional researcher (from PingER Current Member) to guide and advise. The Summer School activities will include: Keynote talk, Presentations on current projects related to Pinger that open up challenges for students to solve and address, and case studies (assignments based on the presentations) for each team of students.
Anjum plans to be at UM by January 28th, we all hope he has his visa by then.
Renan
Renan has contacted the admin (Vinicius) for the UFRJ PingER monitor, to see if it can be restored to a working state. Vinicius explained they went through infrastructure changes and there are some important hardware pieces missing, which is delaying the process. Vinicius explained that hopefully around mid-February they should have the necessary hardware so they can reinstall the systems and PingER program.
Renan has provided a 4 page Appendix on PingERLOD to the ICFA report. This is also available at PingER LOD Overview
Renan is testing Virtuoso which is free to see if it can overcome the slowness of the current PingERLOD. He plans to be done in the next 2 weeks. If this does not work we may need to look into a tool that costs money.
Renan is looking at submitting a paper or poster for a Big Data and Social Computing conference that will be held at Stanford: He is pretty sure he can argue that we have made nice contributions for social computing using (big) data handling technologies. He will discuss with his professor at Rio so she can help him with advice. Bebo will look at the conference information and the deadline to verify its relevance. Renan will look into whether he can get financial assistance to attend.
It would be good if somebody else was able to work with Renan to move PingER LOD forward. However we think the performance issue has to be resolved first.
UM
Ridzuan was unable to attend but sent the following status report: "My progress is that currently I am setting up the Hadoop environment to our UM servers. There have been some installation problems and will take some time to debug and testing Pinger data. We will start using Malaysia pinger data first as a starter. Will inform you the progress in our next meeting."
Badrul plans for his undergraduate student (Abdulrahim Haroun Ali) to have a draft paper on anomalies in PingER measurements to share for review by the end of semester (January).
UM appears to be experiencing very high packet losses (>4%). We need help in understanding these. It does not look like simple congestion since the jitter and RTT is staying low. Saqib needs assistance from Badrul as part of his case study of Malaysia. Badrul writes: "UM is now currently changing the new policy regarding the bandwidth and connection. They are upgrading the switch and router to a new router - mostly every weekend they shut down the network. therefore we are now having some difficulty." Saqib has identified what appear to be losses starting at a JARING router in KL. Badrul will run some mtr traceorute/pings to various sites.
UNIMAS
The tool to enable synchronizing Malaysian monitors is completed. It provides the ability to add, edit and delete nodes. Johari has sent less the URLs to try out. Les has not had much time so far only looked at add form made some suggestions, and edit form - looks good. Johari is working on authentication for the tool. Once that is ready he will send to Badrul and Saqib.
The traceroute server at http://pinger2.unimas.my/cgi-bin/traceroute.pl has the same problem as before. They know (sort of) the problem but haven't got the chance to rectify it (mapping NAT address, needs to be added). There is no progress 12/4/2013, 1/8/2013, 1.22.2014.
Johari has a research student who finalized a proposal in order to officially apply for his masters. He will start in February. He is currently working on threshold/anomaly detection, and will extend to correlating performance over multiple routes.
UTM
Saqib has finished his thesis. He is applying for a Postgraduate Research assistant-ship at UTM where he plans to work on PingER. He is still waiting to hear the result. Update?
Saqib is working with the IT department in UTM to solve the problems of delays in traceroute. They have a new ISP. Now the pinger server is working with single static IP (161.139.68.188) both for intranet and internet. Previous internet IP 161.139.146.158 is removed.
- No load balancing is involved as UTM is working with single ISP.
- Still the problem of the delay in traceroute is not solved. It appears to be due to the DNS lookup delay.
Saqib has started a case study of Malaysian hosts seen from Malaysian Monitors. the case study is identifying the need for the Malaysian monitors to collect traceroutes to their target hosts daily. SLAC has a script to facilitate this, for routes seen from SLAC, that could be adapted for other sites. The latest copy of the word file was shared with Badrul and Johari.
UUM
Dec 4, 2013: The director of computer centre has agreed to support this collaboration and already appointed one staff to be with us during the configuration. Next step, Adib needs to discuss with Prof. Suhaidi to dedicate one PC OR get support from Badrul grant to buy one PC for this purpose or have a bake sale and buy a Raspberry Pi. Adib will do his best to setup PingER monitoring host at UUM before the end of this year. Adib was not on the meeting. He provided an update by email: "Still waiting to get PC. But they promise to provide one very soon. it is totally beyond my control." Regarding the meeting, Adib will attend the upcoming meetings once UUM pinger host is up.
NUST
At the Connect Asia Pacific Summit in Bangkok 2 weeks ago and seeing the project "Mapping the pan Asia Pacific information Superhighway and closing gaps in infrastructure connectivity" Shahryar found that very much related to the work in the PingER project. So Shahryar sent email to a UN agency for a possible collaboration with them on PingER project. He has heard nothing so he will write a detailed proposal and then should contact them again.
Hassan reports that they have established contact with people in Bahawalpur and also waiting to hear from people in Sahiwal. They are very hopeful that link/node shall be up in Bahawalpur very soon.
We are now gathering data from Pingerisl-air and Quest
We are still unable to gather data from airuniversity, buitms, cae, duhs, pingerisl-air, pingerisl-qau, quest, uaf.
Monitors Status
Airuniversity New installation will be install in a week
Buitms Problem in hardware, they are trying to fix it
Cae Problem with IP, they are trying to resolve it
Duhs They are purchasing new servers in two week time
Pingerisl-qau Contact person is still on live
Uaf Problem in system, visit approval for Faisalabad University in progress
In addition we cannot gather data from sau.seecs.edu.pk for all of January and have been unable to gather data from pinger.pern.edu.pk since Jan 13. So we recovered 2 and lost 2.
TULIP - Raja
We have sent a reminder email to Yahoo giving them an update on TULIP and how it might be extended to meet their needs. They have responded, they appear to be interested. We are trying to interest them in providing landmarks.
Raja uses the options feature of traceroute.pl to reduce the time for the PingER/perfSONAR landmarks to make the pings by a factor of 5. Given this we probably need to review the timeouts. Also we are now using caching and nightlty we populate the caches by a cronjob tracerouting to the PingER targets. THis also help speed things up.
Raja has used TULIP's geolocation to provide a Visual traceroute. It can be accessed via http://www.slac.stanford.edu/comp/net/tulip/. It also shows the route as seen GeoIPTools and other methods. Several features have been added to speed it up: the caching, and not bothering with regions where there are insufficient landmarks etc.
PingER at SLAC
The annual PingER report for the International Committee on Future Accelerators (ICFA) is completed. It is available at http://www.slac.stanford.edu/xorg/icfa/icfa-net-paper-jan14/report-jan14.docx.
Raja has also re-written the historical traceroute facility that takes nightly traceroutes and allows one to retrieve and compare them. He then extended it to use the reverse traceroute facility to provide histories for NUST, CERN, UM as well as SLAC. When the UNIMAS and UTM traceroutes work, we can include them. See for example http://www-wanmon.slac.stanford.edu/cgi-wrap/traceroutearchive.cgi?from=maggie1.seecs.edu.pk&to=pinger.cern.ch&date1=2014_01_29&date2=2014_01_30&date3=2014_01_31.
Old Items
TULIP
The new beta test TULIP site is up and running and is at http://tulip.slac.stanford.edu. To first order (due to the number of landmarks available) it only works in N. America, Europe and Pakistan. Even then it is only accurate to a hundred or so km. It also will not work for targets that do not respond to pings or are connected via geo-stationary satellites. Its main use at the moment maybe to find roughly the location, i.e. region/country/state, a target is in. This is particularly useful for proxies and for routers (the latter are typically mis-found by GeoIPtools to be in the corporate HQ of the owner (e.g. Berkeley for ESnet routers). It would be really valuable if router owners provided DNS LOC records filled out.
There was interest in TULIP from Mridul Jain <mridul@yahoo-inc.com>and the Senior Architect in Yahoo! R&D Software Dev in Bangalore, India. He pointed to an interesting paper on Geolocation using CBG and then finding hosts in the area and getting their area codes and using a virtual landmarks making traceroutes. The paper is at https://www.usenix.org/legacy/event/nsdi11/tech/full_papers/Wang_Yong.pdf. He wants something to use in production, however I do not think we are ready.
Linked Open Data
Renan finished the new pingerlod web site. The new thing is that it should be much easier now to modify the info texts. What Renan did was to put the texts into a separate file. The new version has been loaded on the server and some text added to describe how to use the map. However there is a bug that prevents it from executing the map. Renan reports that the bugs should be easy to fix. He has talked to his professor who suggested trying RDF Owlink, it should have faster responeses to queries. Renan will research this. It will probably mean reloading the PingER data so is a lot of work, hopefully this will improve performance. Before the rebuild he will make the fixes and provide a new WAR for us to load on pingerlod.slac.stanford.edu. He is also working on documentation (he has finished the ontology and has a nice interactive tool for visualizing it, since the ontology is the core of the data model of our semantic solution, this will be very helpful for anyone who uses our system, both a developer of the system and a possible user) and his thesis. Bebo pointed out that to get publicity and for people to know about the data, we will need to add pingerlod to lod.org.
Things he will soon do regarding documentation:
- A task/process flow writing all java classes involved on all those batch jobs;
- A Javadoc <http://www.oracle.com/technetwork/java/javase/documentation/index-jsp-135444.html> which will explain all classes and how they are used.
For the Linked Open Data / RDF which is in pre-alpha days, you can go to http://pingerlod.slac.stanford.edu. As can be seen this page is not ready for prime time. However the demos work as long as one carefully elects what to look at:
- Click on Visualizations, there are two choices:
- Multiple Network Metrics: Click on the image: gives a form, choose from Node pinger.slac.stanford.edu pinging to www.ihep.ac.cn, time parameters yearly, 2006 2012, metrics throughput, Average RTT Packet loss and display format Plot graph, then click on submit. In a few seconds time series graph should come up. Mouse over to see details of values at each x value (year).
- A mashup of network metrics x university metrics Click on image: gives another form, pinging from pinger.slac.stanford.edu, School metric number of students, time metric years 2006 2012, display format plot graph, click on submit. Longer wait, after about 35 seconds a google map should show up. Click on "Click for help." Area of dots = number of students, darkness of dots = throughput (lighter is better), inscribing circle color gives university type (public, private etc.) Click on circle for information on university etc.
- Renan will be working on providing documentation on the programs, in particular the install guide for the repository and web site etc. This will assist the person who takes this over.
Renan is using OWLIM as RDF Repository. He is using an evaluation version right now. Renan looked into the price for OWLIM (that excellent RDF Database Management System he told us about). It would cost 1200EUR minimum (~ 1620 USD, according to Google's rate for today) for a one time eternal license. It seems too expensive. No wonder it is so good. Anyhow, he heard about a different free alternative. Just not sure how good it would be for our PingER data. He will try it out and evaluate. He will also get a new evaluation of the free OWLIM lite.
He has also made some modifications on the ontology of the project (under supervision of his professor in Rio) hence he will have to modify the code to load the data accordingly.
Raspberry Pi
A quick comparison of the performance of the two hosts (raspberry pi and regular UNIMAS host) without statistical quantification is available at https://confluence.slac.stanford.edu/display/IEPM/Comparison+of+PinGER+RTTs+from+UNIMAS+monitors+N4+and+RASPBERRY. A page has been created to compare the hardware spec between the pinger.unimas.my node (Intel architecture) and the pinger2.unimas.my node (Raspberry Pi ARM architecture), available from the unimas pinger website at http://pinger.unimas.my/pinger/hardware.php. There is a link to hardware.php in the Comparison+of+PinGER+RTTs+from+UNIMAS+monitors+N4+and+RASPBERRY web page.
Tulip
We are looking at submitting a paper to IMC 2014. April 30, 2014 (see http://conferences2.sigcomm.org/imc/2014/cfp.html) is the submission date. this would discuss how TULIP works and share our experiences. Thus we might focus on: managing the appearance/disappearance of landmarks; the selecting of optimum timeouts for speedup plus any other ideas for speedup; where one could go next (e.g. colocation of landmarks on Yahoo, Google, Hotmail ... sites; impacts of heavy use and scaling (e.g. need for cluster for reflector, multiple landmarks at sites to not overwhelm a given landmark); concerns about network utilization); the relation of accuracy to landmark density (big item). Also more on potential uses such as visual traceroute, identifying proxies. Also maybe a bit more on the actual user interface (results from multiple sources besides TULIP) and implementation (parallel threads, tiering and adding new regions). The paper would assume we have a working TULIP as described in the other paper.
Follow up from workshop
- Hossein Javedani of UTM is interested in anomalous event detection with PingER data. Information on this is available at https://confluence.slac.stanford.edu/display/IEPM/Event+Detection. We have sent him a couple of papers and how to access the PingER data. Hossein and Badrul have been put in contact. Is there an update Badrul?
The Next step in funding is to go for bigger research funding, such as LRGS or eScience. Such proposals must lead to publications in high quality journals. They will need an infrastructure such as the one we are building. We can use the upcoming workshop (1 specific session) to brainstorm and come up with such proposal. We need to do some groundwork before that as well. Johari will take the lead in putting together 1/2 page descriptions of the potential research projects.
- Need to identify a few key areas of research related to PingER Malaysia Initiative and this can be shared/publicized through the website. These might include using the infrastructure and data for: anomaly detection; correlation of performance across multiple routes; and for GeoLocation. Future projects as Les listed in Confluence herehttps://confluence.slac.stanford.edu/display/IEPM/Future+Projects can also be a good start and also Bebo's suggestion.
- Need to synchronize and share research proposals so as not to duplicate research works. how to share? Maybe not through the website, or maybe can create a member only section of the website to share sensitive data such as research proposal?
Anjum suggested Saqib, Badrul and Johari put together a paper on user experiences with using the Internet in Malaysia as seen from Malaysian universities. In particular round trip time, losses, jitter, reliability, routing/peering, in particular anomalies, and the impact on VoIP, throughput etc. It would be good to engage someone from MYREN.
Potential projects
Future meeting - Les
Next meeting Wednesday February 19th 2014 8:00pm Pacific Standard Time, Thursday February 20th 2014 9:00am Pakistan time, Thursday February 20th, 2014 noon Malaysian time, Thursday 20th February, 2014 02:00am Rio Standard Time.