Teng Ma

Ph.D

Innovative Computing Laboratory(ICL)
Computer Science Department
University of Tennessee, Knoxville
1122 Volunteer Blvd
Knoxville TN 37996-3450

Office: 309 Claxton
tma AT eecs DOT utk DOT edu

RESUME

Bio

Teng Ma was born in Nanjing, China. He received his high school education from the Nanjing No.12 High School. He obtained a Bachelor's degree and a Master's degree in Computer Science and Engineering from Southeast University in Nanjing, China, in 2003 and 2006, respectively. In 2006, he began his PH.D. study at the Computer Science Department at the University of Tennessee, Knoxville. Meanwhile, he worked as a Research Assistant at the Innovative Computing Laboratory (ICL) under the guidance of Dr. Jack Dongarra and Dr. George Bosilca. His research interests focused on communication libraries, e.g., Message Passing Interface (MPI). While at ICL, he was an active developer in the Open MPI project. His research interests focused on high performance network/interconnect, programming models, message passing and RDMA. His paper \HierKNEM: An Adaptive Framework for Kernel-Assisted and Topology-Aware Collective Communications on Many-core Clusters" was awarded the best paper award in the 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS '12). He also actively participated in collaborative research projects. He completed a summer internship at Los Alamos National Laboratory in 2009 and another at NetApp Inc. in 2012. Teng Ma was also very active at serving in the high performance computing communities: he served as a student volunteer in ACM/IEEE Supercomputing conferences from 2008 to 2011. He graduated and got his PH.D degree in 2012. Now, He works as a technical staff member in NetApp Inc.

Education

2009 - 2012, Computer Science, University of Tennessee, PH.D, Advisor: Jack Dongarra and George Bosilca

2006 - 2009, Computer Science, University of Tennessee, Master, Advisor: Jack Dongarra and George Bosilca

2003 - 2006, Computer Engineer, Southeast University, Master

1999 - 2003, Computer Engineer, Southeast University, Ba.

Research Summary

Open MPI

Kernel-assisted Intra-node Collective component: KNEM Collective
Kernel-assisted Hierarchical Collective component: HierKNEM Collective
Specific networking supporting for Open MPI's point-to-point communication: ELAN BTL for Quadrics Networks, SICORTEX BTL, SM/KNEM BTL(kernel-assisted Inter-Process Communication approach)

Cell Message Layer(CML)

Develop non-blocking communication libraries for cell messaging layer, which is an MPI-style lightweight communication libraries for LANL's supercomputer `roadrunner'.

FT-MPI

FT-MPI has been developed in the frame of the HARNESS project. The goal of FT-MPI is to provide the end-user a communication library providing an MPI API, which benefits from the fault-tolerance in the HARNESS system. Therefore, FT-MPI implements the whole MPI-1.2 specification, some parts of the MPI-2 document and extends some of the semantics of MPI for giving the application the possibility to recover from failed processes. FT-MPI survives the crash of n-1 processes in a n-process job, and, if required, can respawn them. However, it is still the responsebility of the application to recover the data-structures and the data on the crahsed processes.

DAGuE

DAGuE aims at enabling scientific computing on large scale distributed environments featuring many cores, accelerators and high speed networks. The framework includes libraries, a runtime system, and development tools to help application developers tackle the difficult task of porting their applications to highly heterogeneous and diverse environment. Current and future computing environments leverage highly parallel and heterogeneous hardware systems. Taking advantage of the hardware parallelism exposed by these environments requires highly technical approaches, involving synergies between multiple divergent programming models. This daunting task is critical to expose the hardware parallelism to the application, but due to the involved complexity it remains a solution that only a few programmers can reasonably approach. This project goes out of the sequential, multi-programming mainstream model by proposing a data-flow approach. The application describes a set of data dependencies between tasks, and the DAGuE runtime manages the data transfers and copies in order to complete the execution of the application. It seamlessly integrate accelerators into the execution environment, allowing applications to portably maximize their efficiency on any heterogeneous distributed computing environment.

PH.D Dissertation

Teng Ma Kernel-assisted and Topology-aware MPI Collective Communication among Multicore or Many-core Clusters,

Selected Publications

Teng Ma, George Bosilca, Aurelien Bouteiller, and Jack J. Dongarra, Kernel-assisted and topology-aware MPI collective communications on multi-core/many-core platforms, Journal of Parallel and Distributed Computing, available on online and accepted Manuscript.

Teng Ma, George Bosilca, Aurelien Bouteiller, and Jack J. Dongarra, HierKNEM: An Adaptive Framework for Kernel-Assisted and Topology-Aware Collective Communications on Many-core Clusters, 26th IEEE International Parallel and Distributed Processing Symposium(IPDPS '12),Shanghai, China, May 2012.(Best Paper Award) Accept Ratio: 118/569 = 20%

Teng Ma Kernel-assisted MPI Collective Communication among Many-core Clusters, The 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing(CCGRID12), Doctoral Symposium, Ottawa, Canada, May 2012.

Teng Ma, Aurelien Bouteiller, George Bosilca and Jack J. Dongarra, Impact of Kernel-Assisted MPI Communication over Scientific Applications: CPMD and FFTW, 18th EuroMPI(2011) conference in Santorini, Greece..

Teng Ma, Thomas Herault, George Bosilca and Jack J. Dongarra, Process Distance-aware Adaptive MPI Collective Communications, IEEE Int'l Conference on Cluster Computing (Cluster 2011), Sept. 2011, Austin, Texas. Accept Ratio: 39/140=28%.

Teng Ma, George Bosilca, Aurelien Bouteiller, Brice Goglin, Jeffrey M. Squyres and Jack J. Dongarra, Kernel Assisted Collective Intra-node MPI Communication Among Multi-core and Many-core CPUs, Int'l Conference on Parallel Processing (ICPP '11), Sept. 2011, taipei, taiwan. Accept Ratio: 81/363= 22%

Teng Ma, George Bosilca, Aurelien Bouteiller and Jack J. Dongarra, Locality and Topology Aware Intra-node Communication among Multicore CPUs, 17th EuroMPI(2010) conference in Stuttgart, German, PP 265-274.

Teng Ma and Junzhou Luo: A Prediction-Based and Cost-Based Replica Replacement Algorithm Research and Simulation., AINA 2005: 935-940

Teng Ma, Ye Zhu and Junzhou Luo: A SOA-oriented e-learning resource integration infrastructure and its service discovery model., CSCWD (2) 2005: 1018-1023

Teng Ma and Junzhou Luo: Optimizing Large File Transfer on Data Grid., GCC 2005: 455-460

Xiaopeng Wang, Junzhou Luo, Aibo Song and Teng Ma: Semantic Access Control in Grid Computing., ICPADS (1) 2005: 661-667

Junzhou Luo, Aibo Song, Ye Zhu, Xiaopeng Wang, Teng Ma, Zhiang Wu, Yaobin Xu and Liang Ge: Grid Supporting Platform for AMS Data Processing., ISPA Workshops 2005: 276-285

Feng Li, Junzhou Luo, Yinying Yang, Ye Zhu and Teng Ma: A Dynamic Grid Authorization Mechanism with Result Feedback., GCC 2004: 867-870

Ye Zhu, Junzhou Luo and Teng Ma: Dividing Grid Service Discovery into 2-Stage Matchmaking., ISPA 2004: 372-381

Recent Talks

"HierKNEM: An Adaptive Framework for Kernel-Assisted and Topology-Aware Collective Communications on Many-core Clusters", Invited talk at CAS ICT(Chinese Academic of Science, Institution of Computing and Technology) , Beijing, China, June-5-2012.

"HierKNEM: An Adaptive Framework for Kernel-Assisted and Topology-Aware Collective Communications on Many-core Clusters", Invited talk at CAS SIAT (Chinese Academic of Science, Shenzhen institution of Advance Technology), Shenzhen, China, June-4-2012.

"HierKNEM: An Adaptive Framework for Kernel-Assisted and Topology-Aware Collective Communications on Many-core Clusters", IPDPS 2012, Plentary Talk (Best Paper), Shanghai, China, May-24-2012. Vedio

"Kernel-assisted MPI Collective Communication among Many-core Clusters", CCGRID2012 Doctor Forum, Ottawa, Canada, May-14-2012.

"Kernel-assisted and Topology-aware MPI Collective Communication amongãcore or Many-core Clusters", NPH.D disertation proposal, Knoxville, TN, March-26-2012.

"Kernel-assisted and Topology-aware MPI Collective Communication amongãcore or Many-core Clusters", NEC Lab, U.S, Invited Talk, Princeton, NY, March-28-2012.

"Impact of Kernel-asissted MPI Communication on Scientific Applications", SIAM PP 2012, Invited Talk, Savannah, GA, Feb-15-2012.

"HierKNEM:an adaptive Framework for Kernel-assisted and Topology-aware MPI collective on Many-core Cluster", Innovative computing lab, lunch talk, Knoxville, TN, Dec-2-2011.

"Process Distance-aware Adaptive MPI Collective Communications", IEEE Int'l Conference on Cluster Computing (Cluster 2011), Sept. 2011, Austin, Texas.

"Smart MPI Intra-node Communication among Multicore and Manycore Machines", IEEE/ACM Supercomputing conference: China HPC Workshop 2010(ATIP), Nov. 2010, NO, LA.

"Kernel-assisted MPI Communication on Multi-core Clusters", Innovative computing lab, lunch talk, Knoxville, TN, July-15-2011.

"A Kernel-assisted Topology-aware Scalable Collective Framework", Innovative computing lab, annual retreat meeting, Gatlinburg, TN, Aug-11-2011.

"The latest progress in Open-MPI point-to-point communication(sicortex btl, sm-ioat btl)",Innovative computing lab, annual retreat meeting, Townsend, TN, Aug-15-2009.

"Design and implementation of non-blocking cell message layer communication",Los Alamos National lab 2008 summer student poster, Los alamos, NM, Aug-10-2008.

"New Progress in Open MPI p2p communication: Elan and Sicortex", Innovative computing lab, annual retreat meeting, Townsend, TN, Aug-2008.

"SiCortex walkthrough", Innovative computing lab, lunch talk, Knoxville, TN, Dec-12-2008.

Professional Service

Program Committee Member of The Fourth International Workshop on Data Intensive Computing in the Clouds (DataCloud) 2013

Program Committee Member of 2013 International Conference on Parallel Processing (ICPP-2013)

Program Committee Member of EuroMPI 2013

reviewer of journal of cloud computing

reviewer of journal of Grid Computing

reviewer of IPDPS 2013 workshop HPPAC'13

reviewer of 17th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP)

Teng Ma
Last modified on Feb 20, 2013