Time & date
Wednesday July 15th 2015 10:00pm Pacific Standard Time, Thursday July 16th 2015 10:00am Pakistan time, Thursday July 16th 2015 1 pm Malaysian time, Thursday July 16th, 2015 02:00am Rio Standard Time.
Coordinates of team members:
See: http://pinger.unimas.my/pinger/contact.php
Attendees
Invitees:
Hassaan Khaliq, Kashif, Raja, Samad Riaz (SEECS); Johari-, Nara, Adnan Khan (UNIMAS); Abdullah, Badrul, Anjum+, Ridzuan?, Ibrahim? (UM); Hanan, Saqib (UTM)+; Adib+, Fatima+ (UUM); Fizi Jalil (MYREN); Thiago+, Les+, Bebo+ (SLAC)
+ Confirmed attendance
- Responded but Unable to attend:
? Individual emails sent
Actual attendees:
Anjum, Adib, Fatima, Maria, Thiago, Les, Bebo,
Topic
The topic for this Skype meeting is plans to submit papers to NETAPPS15 (http://netapps2015.internetworks.my/v2/)
The full paper submission is July 31st, 2015 (already extended once and now a firm deadline). Papers are a max of 6 pages including figs and refs. The call for papers is at http://netapps2015.internetworks.my/v2/index.php/call-for-papers. There is information on the submission at http://netapps2015.internetworks.my/v2/index.php/paper-submission. It will be in Putrajaya 1-3 December 2015 and hosted by UUM (Adib is an organizer).
If there were sufficient papers concerning PingER than maybe there would be a separate session for PingER.
Raspberry Pi & PingER Thiago, Les, Bebo, Johari
There is interest and work in progress for SLAC on "Worldwide Internet performance measurements using lightweight measurement platforms". This will be based on PingER on a Raspberry Pi, see ePinger Project at SLAC. I doubt neither Thiago nor I can get the travel funds. Johari can present at the meeting. Bebo will contact Topher to see if he can put something together on potential Android implmentation
Malaysian case study: Saqib, Anjum
We tried calling Saqib several times but with no success. Anjum will try and contact after the meeting.
GeoLocation: Main author(s): Anjum; others contributing Raja, Les
Anjum is looking at putting together a paper on the TULIP geolocation as applied to Pakistan, where it is accurate to +- 10km.
Big Data Analysis: Main authors: Thiago, Renan, other contributors Maria, Les...
Abstract: Perhaps the new technologies, managing big volumes of data still is a complex task and demands great efforts. This work has as its objective applying existing open-source technologies capable of dealing with a huge volume of data with OLAP characteristics in an efficiently manner. The concepts and specific technologies of an analytic scenario will be applied over data from a quality and performance internet measurements dataset. The article will contemplate all the process, from the data extraction in its origin until the data analysis using OLAP queries, passing through the data transformation process and the technologies used to build the environment.
Thiago is trying to get the Hadoop cluster set up at SLAC He is working with Karl Amrhein, who has been on vacation. Thiago has it running on his own laptop at the moment.
Even though Rean is busy with other matters, including his MSc thesis final presentation on July 23rd, he is willing to closely collaborate with Thiago. he cannot just build up the paper from scratch myself, but he can definitely closely support Thiago on writing it. By the way the thesis is about scientific big data processing in large clusters (over 1000 cores), enabling data management & analyses of a large volume of data flowing in a complex dataflow. It will also be in English.
Others
- Maria noted that UFRJ has problems with infrastructure. they have transformed the original PingEr data into triplets. Maria will contact UFRJ (Cristiane) to see whether to propose something, maybe something to do with the bloat in data size. Maria will send out an email today and let us know by the weekend. From previous report:Ibrahim has not nothing significant yet.
- Maria Luiza has requested Raphaela and Christiane to give an update on what they've been working on in the last weeks, and maybe talk with Thiago on how they've been using their local cloud of 4 nodes with Cloudera and a data cube version of the ontology.
- Cristiane reports (7/1/2015): "I am trying to automatize the triplification of PingER data on Kettle. For now, part of the transformation is made on Kettle and another is made by a Java code. Although this solution works for a data sample, is important to have the entire process on Kettle because it facilitates to understand, modify and control the triplification process."
- Fatima is putting things together, she has Hadoop on 2 machines at the moment. There are no results yet, she will know by end of next week (24th July) whether there is sufficient content for a paper.
Summary
It is currently unclear how many papers can be created in the time available. Even if created with the rush, they may not be complete enough to be accepted. A step back from a PingER session might be to have a session including PingER and other related papers.