...
Attendees
Invitees:
Hassaan Khaliq Khaliq? (SEECS); Saqib+, Aqsa (UAF); Johari+-, Adnan Khan (UNIMAS); Badrul, Ridzuan, Ibrahim (UM); (UTM); Adib?, Fatima (UUM); Fizi Jalil (MYREN); Les+, Bebo-, Joao+ (SLAC).
...
? Individual emails sent
Actual attendees:
Saqib, Les
Administration
Joao Lucas Rulff Da Costa, Brazillian student: arrived at SLAC 5/16/2016. Here until end August 2016.
Need to add to http://pinger.unimas.my/pinger/contact.php. Joao, please provide your coordinates to Johari. jolucasrulff is joaolucasrdacosta is Skype ID.
Added jlrulff@gmail.com to pinger-my Google group
Will be working on making up-to-date PingER data available for download via FTP
Android - Bebo
No update 5/18/2016
Bebo has set up a Github codebase as a new project. It contains the PingER MA (pinger2.pl and the traceroute/ping server). Anyone needs to sign up for a Github account (if you don't already have one), so you can be added as a project member.
- slac-pinger/pinger created by topherwhite. PingER project https://github.com/slac-pinger/pinger
- Now we have it we can share with Amity to check it works.
- Bebo reports that someone from Amity requested to be a member of the Android group.
UUM
No update 5/18/2016
From Adib:
"Since we're still waiting on an account, at SLAC, I asked the student to explore another idea (which is quite relevant) proposed by Prof. Bebo “Creative visualization of PingER data, including rich interaction;”. He has already started looking at the possible attractive/interactive design, at the same time he is working to improve his programming skills."
"On the other hand, I am in the final stage of writing a case study on the Internet performance in ASEAN countries and its implication on the Socio-Economic Indexes. Will try to share with Dr.Les and Prof. Bebo soon to comment and get your advice on the publication possibility of the paper."
UNIMAS
No update 5/18/2016
...
UAF (Saqib)
Name: Aqsa Hameed
Title: visualization of PingER historical data using warehouse
...
There are 4 students: Aqsa Hameed, Saba Muzamil, Tahseen and Sara Masood. They are all busy this week and last week with exams. We can expect more updates in a couple of weeks time when the exams are done.
- Aqsa and Saba are working together. Their goal is focusing on visualization of PingER historical data using warehouse. The idea is to develop a warehouse in
...
- UAF university and make it publicly available.
...
- They are 50-60% done with setting up a Hadoop cluster with 3 nodes, 1 master, 2 slaves. She is currently working on importing the PingER data into hdfs on the cluster. they have run some Impala queries on the data and are working on visualization
- Tahseen is working on Handling Missing Data in PingER by using Artificial Intelligence. He has executed some Java queries and looking at parallel scientific workflow management using SciCumulus.
- Sara Masood is working on a PingER Internet Performance Monitoring Agent On Android Device. There has been a discussion to pin down more closely what is needed, i.e. a PingER Measurement Agent running on an Android making measurements to the Beacons, and the ability to upload the data nightly. We held a skype group meeting with Topher, Sara, Saqib, Bebo and Les to move forward. See 20160331 Meeting on PingER/Android, also see Discussion on porting PingER to Android. She has what she needs for now.
Project on Internet performance has been accepted, but no funding yet.
UTM
Saqib's old supervisor agreed to appoint a master student to take of PingER in UTM. Saqib has emailed 3/9/2016, no progress 4/6/2016, no progress 5/18/2016..
UM
MYREN
pingersonar-utm.myren.net.my is down. Emailed Fitzi 5/12/2016.
NUST
No update 5/18/2016
Hassaan's student (Mian Anas Abrar <anas.abrar@gmail.com>) reports: "Sir haven't disabled any of them which i told earlier and actually can't even get back to them because of their wrong contact numbers or their numbers not responding since then. I have tried contacting Dr.Les earlier few time as per your instructions but don't know why i wasn't able to email him. rest the status is actually same nodes being down". We believe we have solved the mail problem.
We are unable to gather data from the following Pakistani PingER monitors:
- aup.seecs.edu.pk
- cae.seecs.edu.pk
- maggie1.seecs.edu.pk
- maggie2.seecs.edu.pk
- npm.neduet.edu.pk*
- pinger.iba-suk.edu.pk host is pingable
- pinger.isra.edu.pk host is pingable
- pinger.uettaxila.edu.pk*
- pingerisl-air.pern.edu.pk
- pingerisl-qau.pern.edu.pk*
- pingerjms.pern.edu.pk
- pingerkhi-uok.pern.edu.pk
- pingerlhr-pu.pern.edu.pk* host is pingable
- pingerpwr.pern.edu.pk* host is pingable
- pingerqta.pern.edu.pk
- quest.seecs.edu.pk*
* This hosts was noted at the 3/9/2016 meeting as being down for the last 30 days.
Dr Hassaan Khaliq contacted and said "We shall look into the issues with them" 5/13/2016.
PingER at SLAC
Working on the following hosts to be able to gather data
| Host | State | last seen | Status |
|---|---|---|---|
| |||
| pinger.arn.dz | email 1/30/2016, 2/22/2016, 4/11/2016 no response. Giving up. | Nov 2015 | Does not ping |
| pinger.unimas.my | email 5/12/2016 | March 2016 | Does not ping |
| pingersonar-utm.myren.net.my | email to Fitzi 5/12/2016 | Does not ping |
Next Meeting
Next meeting: Wednesday Jun 22nd 9:00pm Pacific Standard Time, Thursday June 23rd, 2016 9:00am Pakistan time, Thursday Jun 23rd 2016 12:00noon Malaysian time, Thursday June 23rd 2016 02:00am Rio Standard Time.
Old Items
Visualization ideas for PingER moved here 5/20/2016
- One way of drilling down is via a map of all sites. Faisal Zahid put something together a couple of years ago. It is at http://www.slac.stanford.edu/comp/net/wan-mon/viper/pinger-coverage-gmap.html. It uses Google maps, allows pan and zoom, plus selection of type (measurement agent, beacon, target) of host to display. It also allows one to select a host (Search location), mouse click on the balloon to find out more about the host, from there click on the sites web site, click on the country or region to get the relevant pingtable.pl information. The Visualize allows one to choose a measurement agent and a country and it draws coloured lines between the sites to represent the metric choosen. This was very useful but parts of it are broken. It would be great to fix or replace. It has some documentation.
- For plotting time series, Grafana looks very promising
- Updating the motion charts, adding newer data etc. would be most useful. See http://www-iepm.slac.stanford.edu/pinger/pinger-metrics-motion-chart.html and http://www-iepm.slac.stanford.edu/pinger/explorer.html
- Enhance the selection Metrics, Ticks (see pingtable.pl) to allow multiple metrics (min, avg, max RTT, loss, thrpughput, , MOS ..) to be plotted as a time series.
- Enhanced selection remote hosts for selected tick, selected metric selected mutiple targets (pingtable.pl currently only allows one target), e.g. Once you have plot enable fitting the data to timelines (linear, exponential, power series)
- Map pf countries the world colored by metric allow selection of metric and for a particular year. with chosen MA,
For plotting time series, Grafana looks very promising
Updating the motion charts, adding newer data etc. would be most useful. See http://www-iepm.slac.stanford.edu/pinger/pinger-metrics-motion-chart.html and http://www-iepm.slac.stanford.edu/pinger/explorer.html
Enhance the selection Metrics, Ticks (see pingtable.pl) to allow multiple metrics (min, avg, max RTT, loss, thrpughput, , MOS ..) to be plotted as a time series.
Enhanced selection remote hosts for selected tick, selected metric selected mutiple targets (pingtable.pl currently only allows one target), e.g.
Once you have plot enable fitting the data to timelines (linear, exponential, power series)
Map pf countries the world colored by metric allow selection of metric and for a particular year. with chosen MA, e.g.
Name: Sara Masood
Title: PingER Internet Performance Monitoring Agent On Android Device
Status: There has been a discussion to pin down more closely what is needed, i.e. a PingER Measurement Agent running on an Android making measurements to the Beacons, and the ability to upload the data nightly. We held a skype group meeting with Topher, Sara, Saqib, Bebo and Les to move forward. See 20160331 Meeting on PingER/Android, also see Discussion on porting PingER to Android. She has what she needs for now.
Name: Saba Muzamil
Title: She is interested to work on Environment that supports the execution of analytical distributed queries in Hadoop Cluster.
Name: Tahseen
Title: Handling Missing Data in PingER by using Artificial Intelligence
Status: He is doing the literature review to find out the best algorithm to handle missing data in PingER.
About IoT:
Title: Air Quality measuring network in University of Agriculture Faisalabad.
Status: Now we have two Raspberry Pi-2 and few air quality measuring sensor i.e.; NO, CO, O3. Currently, working on the configuration and programming of the Pi-2 and the sensors.
Project on Internet performance has been accepted, but no funding yet.
UTM
Saqib's old supervisor agreed to appoint a master student to take of PingER in UTM. Saqib has emailed 3/9/2016, no progress 4/6/2016.
UM
MYREN
pingersonar-utm.myren.net.my is down. Emailed Fitzi 5/12/2016.
NUST
Hassaan's student (Mian Anas Abrar <anas.abrar@gmail.com>) reports: "Sir haven't disabled any of them which i told earlier and actually can't even get back to them because of their wrong contact numbers or their numbers not responding since then. I have tried contacting Dr.Les earlier few time as per your instructions but don't know why i wasn't able to email him. rest the status is actually same nodes being down". We believe we have solved the mail problem.
We are unable to gather data from the following Pakistani PingER monitors:
- aup.seecs.edu.pk
- cae.seecs.edu.pk
- maggie1.seecs.edu.pk
- maggie2.seecs.edu.pk
- npm.neduet.edu.pk*
- pinger.iba-suk.edu.pk host is pingable
- pinger.isra.edu.pk host is pingable
- pinger.uettaxila.edu.pk*
- pingerisl-air.pern.edu.pk
- pingerisl-qau.pern.edu.pk*
- pingerjms.pern.edu.pk
- pingerkhi-uok.pern.edu.pk
- pingerlhr-pu.pern.edu.pk* host is pingable
- pingerpwr.pern.edu.pk* host is pingable
- pingerqta.pern.edu.pk
- quest.seecs.edu.pk*
* This hosts was noted at the 3/9/2016 meeting as being down for the last 30 days.
Dr Hassaan Khaliq contacted and said "We shall look into the issues with them" 5/13/2016.
PingER at SLAC
Preparing a presentation for an NREN conference (BDREN) in Dhaka Bangladesh on 18 April. Will present by video.
Working on the following hosts to be able to gather data
| Host | State | last seen | Status |
|---|---|---|---|
| web.hepgrid.uerj.edu | Worked from April 26th to May 3rd, then not pingable. Sent email 5/12/2016. | Apr 26, 2016 | does not ping |
| pinger.arn.dz | email 1/30/2016, 2/22/2016, 4/11/2016 no response. Giving up. | Nov 2015 | Does not ping |
| pinger.unimas.my | email 5/12/2016 | March 2016 | Does not ping |
| pingersonar-utm.myren.net.my | Does not ping |
Next Meeting
Next meeting: Wednesday Jun 22nd 9:00pm Pacific Standard Time, Thursday June 23rd, 2016 9:00am Pakistan time, Thursday Jun 23rd 2016 12:00noon Malaysian time, Thursday June 23rd 2016 02:00am Rio Standard Time.
Old Items
Ibrahim has extracted the PingER Zip manually. He is reconstructing 11Gb of data, has 15GBytes of data there. He is trying to use SPAR to classify the data. He also was looking at RDF. the next step is to use MapReduce to organize and reduce the output of data so can visualize it. He will be using a the same techniques he used for looking at 1996-2006 weather data. However at the moment he cannot access his VMs from myren for the last two weeks, even for the myren site, He has emailed them but till now they have not fixed the issue. He has updated all his data in their cloud. No update 1/6/2016. Moved here 4/6/2016.
Huffington Post moved here 3/8/2016
...