I am a Ph.D. candidate in the Department of Computer Science at Indiana University Bloomington. My research advisor is Prof. Geoffrey Fox and I work in the Community Grids Lab as a research assistant. In my Ph.D. research, I focus on cloud technologies such as MapReduce, and Dryad. Specifically, I am trying to see how we can improve these technologies by applying them to various data/compute intensive parallel applications and comparing them with traditional implementations such as MPI. I used the knowledge gained to develop a new MapReduce runtime (CGL-MapReduce), which uses a content dissemination network for all its intermediate data transfers and supports iterative MapReduce computations efficiently. In my latest research, I am trying to understand how we can use the above runtimes on virtualized resources and see what performance implications that they would incur on these resources. I am also a committer for Apache Sandesha and Apache Axis2 projects and the PMC for the Apache Sandesha project. You can find the latest information of my research in my CGL-reports blog.
Education
Ph.D. Computer Science (Candidiate) Indiana University BloomingtonM.Sc. Computer Science 2007 Indiana University Bloomington
B.Sc. Engineering 2004 Department of Computer Science and Engineering University of Moratuwa Sri Lanka
Research Interests
Cloud Computing, Data Intensive Scalable Computing, or High Performance Computing
Research
|
CGL-MapReduce - a light weight streaming based map
reduce runtime with the improved support for iterative MapReduce
computations and faster intermediate data transfers. CGL-MapReduce
incorporates the concept of long running processes available
in parallel runtimes such as MPI to the MapReduce model
and it uses NaradaBrokering's
publish/subscribe messaging infrastructure for all
its communication requirements and the intermediate data
transfers. More detailed description of CGL-MapReduce
can be found
here.
Implemented a series of scientific data processing applications such as High Energy Physics data analysis, Cap3 Analysis, Kmeans Clustering, Matrix Multiplication, and Multi-Dimensional Scaling using different parallel runtimes such as Hadoop, Dryad, CGL-MapReduce, and MPI and compared their performance to understand the various benefits and limitations of different parallel runtimes. More detailed description of these solutions and their performance results can be found here. |
|
Researched on the performance
implications of virtualization for High Performance
Computing (HPC) applications, using a private cloud
infrastructure based on Xen and Eucalyptus. This
paper contains the findings of this research.
|
| Integration of Clarens Server, ROOT Analysis Framework and Naradabrokering to provide a collaborative framework for analysing distributed data, especially the data from particle physics experiments. The project is still in working progress. However, a proof of concept implementation/integration is completed. |
| Developed a C++ Client for Naradabrokering to facilitate the integration of Clarens Server from Caltech with Naradabrokering. This eliminates the dependency to have a JVM installed at client machines as in the JNI implementation. |
| Developed a C++ Client for Naradabrokering using JNI technology allowing C++ clients to utilize the publish/subscribe messaging capabilities of Naradabrokering. Software is available for download here. |
| Developed a transport-independent scheme for tracking the availability of entities in distributed systems using Naradabrokering messaging substrate. The scheme enforces the authorized generation and consumption of traces (encapsulating entity availability). |
| Y790 - Performance testing on Multi-core chips (Results). |
| Implemented the message bridge to connect IBM Websphere and Naradabrokering. (The Architecture). |
| Y790 - Common Architecture for Functional Extensions on Top of Apache Axis2 (This was based on the work I did in axis2 project.) |
Publications
| Jaliya Ekanayake, Geoffrey Fox, High Performance Parallel Computing with Clouds and Cloud Technologies Technical Report June 12 2009. |
| Eran Chinthaka, Jaliya Ekanayake, David Leake, CBR Based Workflow Composition Assistant, Accepted for publication, IEEE 2009 Third International Workshop on Scientific Workflows (SWF 2009). |
| Geoffrey Fox, Seung-Hee Bae, Jaliya Ekanayake, Xiaohong Qiu, and Huapeng Yuan, Parallel Data Mining from Multicore to Cloudy Grids, High Performance Computing and Grids workshop, 2008. |
| Shrideep Pallickara, Jaliya Ekanayake, Geoffrey Fox, An Overview of the Granules Runtime for Cloud Computing, Fourth IEEE International Conference on eScience, 2008, pp.412-413. |
| Jaliya Ekanayake and Shrideep Pallickara, MapReduce for Data Intensive Scientific Analysis, Fourth IEEE International Conference on eScience, 2008, pp.277-284. |
| Jaliya Ekanayake, Shrideep Pallickara, and Geoffrey Fox, A collaborative framework for scientific data analysis and visualization, Collaborative Technologies and Systems(CTS), 2008,pp. 339-346. |
| Shrideep Pallickara, Jaliya Ekanayake and Geoffrey Fox, A Scalable Approach for the Secure and Authorized Tracking of the Availability of Entities in Distributed Systems in the proceedings of Proceedings of the 21st IEEE International Parallel & Distributed Processing Symposium (IPDPS 2007). Long Beach, California. |
| Srinath Perera, Chathura Herath, Jaliya Ekanayake, Eran Chinthaka, Ajith Ranabahu, Deepal Jayasinghe, Sanjiva Weerawarana,Glen Daniels Axis, Middleware for Next Generation Web Services on IEEE International Conference on Web Services (ICWS'06) |
| Ajay Smitha and Jaliya Ekanayake, Analysis of the Usage Statistics of Robots Exclusion Standard. In proceedings of the IADIS WWW/Internet 2006 Murcia, Spain 5-8 October 2006. |
| Developerworks Article on Apache Sandesha: Use Apache Sandesha to support Web services implementation. |
| CTS2009 Poster |
eScience08 Posters |
||
 |
 |
 |
|
|
CTS2008 Posters |
|||
|
|
|
||
