Revision as of 11:29, 8 April 2011

Spring 2011 (GF)

Courses: None
Submissions:
- Ranking sponsored online ads (NOSSDAV 11)

working on: Large Scale data processing with MapReduce on GPU/CPU hybrid systems

the report is available here or from this adddress: \students\neshat\reports\large_scale_data_processing\doc\doc.pdf

April 08

worked on Thesis. First version of introduction, background, first and second chapters are ready.

April 01

worked on Thesis, first version of introduction and background are ready.

March 21

Revised NOSSDAV paper

March 14

Worked on software implementation of more advanced version of video advertising. Current software loads keywords from XML file, creates video vector, and load interests from .txt file.
Submitted camera ready version of ICME paper
Prepared presentation for ICME paper

Feb 28

continued to revise predicting quality work. Report is accessible from here.
worked on more advance version of advertising on video. report.
Created a new and updated set of common keywords for 55 different topics.

Feb 15

Started to work on more advance version of advertising on video. report.
revised predicting quality work. Report is accessible from here.

Feb 8

continued to revise predicting quality work. Report is accessible from here.
Went over some papers to find solutions for creating dynamic thread in GPU.

Feb 1

Revised predicting quality work. Report is accessible from here.
Started to implement proposed system for using Hadoop over Hybrid CPU/GPU systems

Jan 24

(On Going)Designing high-level architecture of proposed approach for using Hadoop over Hybrid CPU/GPU systems
Read one example of large scale data proc. with map reduce
Read papers about GPU clusters for HPC
Explored Hadoop and its properties like HDFS
Explored Architecture of NVIDIA GPU cluster's arch and specs

Jan 17

read two papers about Phoenix, a mapreduce implementation for multi-core processors.
spent some days to figure out how to use Mark framework and run some samples, but couldn't fully understand. These works has been done:
- Configured system (windows) to run Mars, including cuda and SDK installation as well as VS9 configuring.
- Corrected some typos in the code (library mismatching)
- Asking authors about problems, and got this answer: "I must apologize that mars_v2 is buggy and complex, and we don't maintain the code base any more, I strongly recommend you to try the latest version on linux"
- tried to install mars_v2 on Linux, but it is still buggy and complex. It seems this frame work could run only with certaing configuration, and with older versions of CUDA.
Explored Mars to find its algorithm, and found in co-processing mode (Hybrid) they partition input data into two parts, one for CPU processing, the other for GPU processing. After the map stage, they merge data on CPU side, then dispatch data again to CPU workers and GPU workers.
Looked at phonix, another System for MapReduce Programming from Stanford. It was the comparison base for Mars.
- Spent 2 days for writing resume and being prepared for YouTube interview.

Jan 10

Explored related works and potential ideas

Fall 2010 (TA)

Courses:
- CMPT-820: Multimedia Systems
- CMPT-825: NLP

worked on:
- effective advertising in video

Submissions:
- SmartAd: a smart autonomous system for effective advertising in video (ICME 11)

Summer 2010 (RA)

- Writing for publication

worked on:
- Estimating the click-through rate for new ads with semantic and feature based similarity

algorithms

Spring 2010 (RA)

Courses:
- CMPT-886: Special topics in operation systems

worked on:
- Accelarting online auction using GPU
- Estimating the click-through rate for new ads with semantic and feature based similarity