From May 25-29, the Institute of Electrical and Electronics Engineers (IEEE) held the 29^th annual International Parallel and Distributed Processing Symposium (IPDPS) in India.

Georgia Tech’s Srinivas Aluru returned to his home city as the invited program chair, a role held last year by the School of Computational Science and Engineering Chair David A. Bader.

Researchers from around the world submitted nearly 500 papers for possible inclusion in the conference agenda. Of the 109 papers that reviewers accepted, Georgia Tech presented six, and one was selected as a “Best Paper.” Graduate student Kenneth Czechowski, and professors Le Song and Richard Vuduc are the Georgia Tech co-authors who received one of the four best paper awards.

The IPDPS conference brings together top international scientists and students to collaborate and delve into many aspects of parallel computing.  They share research and ideas driving the design and performance of leading-edge technology, and expand the set and complexity of problems that can be addressed.

In all, the IPDPS 2015 conference featured 19 workshops, paper presentations, tutorials, and commercial presentations and exhibits.

Attendees from Georgia Tech included:

Leadership

Srinivas Aluru

IPDPS 2015 Program Chair

IPDPS 2015 Steering Committee Member

High Performance Computational Biology (HiCOMB) Workshop Co-chair

David A. Bader

IPDPS 2015 Steering Committee member

High Performance Computational Biology (HiCOMB) Workshop Co-chair

Graph Algorithms Building Blocks (GABB) Workshop Program and Steering Committees Member

Parallel and Distributed Computing for Large Scale Machine Learning and Big Data Analytics (ParLearning) first keynote address titled "Massive-scale analytics"

Heterogeneity in Computing Workshop (HCW) Panelist

Edmond Chow

IPDPS 2015 Program Committee Member

Concettina Guerra

IPDPS 2015 Program Committee Member

Hyesoon Kim

IPDPS 2015 Program Committee Member

Santosh Pande

IPDPS 2015 Program Committee Member

Karsten Schwan

IPDPS 2015 Program Committee Member

Richard Vuduc

High-Performance, Power-Aware Computing (HPPAC) Workshop keynote address titled "On algorithmic execution time, energy, and power"

International Workshop on Automatic Performance Tuning (iWAPT) Steering Committee Member

Best Paper

CA-SVM: Communication-Avoiding Support Vector Machines on Distributed Systems

Yang You (Tsinghua University, China); James Demmel (University of California at Berkeley, USA); Kenneth Czechowski (Georgia Institute of Technology, USA); Le Song (Georgia Institute of Technology, USA); Richard Vuduc (Georgia Institute of Technology, USA)

Papers

A Sparse Direct Solver for Distributed Memory Xeon Phi-accelerated Systems

Piyush Sao (Georgia Institute of Technology, USA); Xing Liu (IBM T.J. Watson Research Center, USA); Richard Vuduc (Georgia Institute of Technology, USA); Xiaoye Li (Lawrence Berkeley National Laboratory, USA)

Grouping Blocks for MapReduce Co-locality

Xiao Yu (Georgia Institute of Technology, USA); Bo Hong (AT&T, USA)

Towards Balance-Affinity Tradeoff in Concurrent Subgraph Traversals

Yinglong Xia (IBM Research, USA); Lifeng Nai (Georgia Institute of Technology, USA); Jui-Hsin Lai (IBM Research, USA) 

Workshop Papers

A Methodology for Co-Location Aware Application Performance Modeling in Multicore Computing

Daniel Dauwe (Colorado State University, USA); Eric Jonardi (Colorado State University, USA);  Ryan Friese (Colorado State University, USA);  Sudeep Pasricha (Colorado State University, USA);  Anthony A. Maciejewski (Colorado State University, USA);  David A. Bader (Georgia Institute of Technology, USA); and Howard Jay Siegel (Colorado State University, USA);

GraphReduce: Large-Scale Graph Analytics on Accelerator-Based HPC Systems

Dipanjan Sengupta (Georgia Institute of Technology, USA);  Kapil Agarwal (Georgia Institute of Technology, USA);  Shuaiwen Leon Song (Pacific Northwest National Laboratory, USA); Karsten Schwan (Georgia Institute of Technology, USA)