Main Page

From HPCRL Wiki
(Difference between revisions)
Jump to: navigation, search
(Publications)
(PhD Students)
(94 intermediate revisions by 15 users not shown)
Line 1: Line 1:
== Tensor Contraction Engine ==
+
== High Performance Computing Research Projects ==
 +
* EAGER - Towards Automated Characterization of the Data-Movement Complexity of Large Scale Analytics Applications.
 +
* XPS: FULL - Collaborative Research: PARAGRAPH: Parallel, Scalable Graph Analytics.
 +
* [[PAdvisor]] - Padding advisor to avoid cache conflict misses.
 +
* [[Whole-Program Adaptive Error Detection and Mitigation]].
 +
* Improving Vectorization.
 +
* Compiler/Runtime Support for Developing Scalable Parallel Multi-Scale Multi-Physics Applications.
 +
----
  
The majority of software for scientific computations is written in the low-level languages FORTRAN and C. The computational structure of some of this software, however, has sufficient underlying structure that it could benefit from special-purpose software engineering tools or domain-specific programming languages. E.g., electronic structure calculations in quantum chemistry and in physics involve large collections of tensor contractions (generalized matrix multiplications). Currently, chemists spend weeks or months manipulating formulas containing dozens or hundreds of terms with Mathematica, hand-optimizing the computation, and writing FORTRAN code by hand. The computation can take on the order of 1 TFLOP week or more and can require multiple TBs of storage.
+
* [[Global Trees]] - Globally addressable distributed tree data structure library and runtime.
 +
* [http://pluto-compiler.sourceforge.net/ Pluto] - An automatic parallelizer and locality optimizer for multicores.
 +
* [[Scioto]] - A scalable taskpools framework for global address space programming models.
 +
* [[Tensor Contraction Engine]] - An optimization system for quantum chemistry tensor equations.
 +
* [[UTS]] - The Unbalanced Tree Search benchmark is an irregular parallel benchmark.
 +
* [[PrimeTile]] - A parametric multi-level tiler for imperfect loop nests.
 +
* [[Hybrid MPI+UPC]] - Hybrid parallel programming with MPI and Unified Parallel C.
 +
* [[Polyhedral Compilation]] - Polyhedral Compilation projects.
 +
* [[DDGVec]] - A dynamic analysis of vectorization potential.
 +
* [[HOSTS]] - The High Order Stencil Transformation Suite: A framework for enhancing data reuse via associative reordering.
 +
* [[SDSL]] - Stencil Domain Specific Language: A language and compiler for developing performance portable stencil code.
 +
* [[IE]] - Inspector/Executor compiler : A source-to-source transformation to target distributed memory systems for partitionable irregular loop computations
 +
* [[SpMV on GPUs]] - Optimizing sparse matrix-vector multiplication on GPUs
  
We are developing a domain-specific language that allows chemists to specify the computation in a high-level Mathematica-style language. The compiler for this language, the '''Tensor Contraction Engine (TCE)''', searches for an optimal implementation and generates FORTRAN code. First, algebraic transformations are used to reduce the number of operations. We then minimize the storage requirements to fit the computation within the disk limits by fusing loops. We have designed an algorithm that finds the optimal evaluation order if intermediate arrays are allocated dynamically and are working on combining loop fusion with dynamic memory allocation. If the computation does not fit within the disk limits, recomputation must be traded off for a reduction in storage requirements. If the target machine is a multi-processor machine, we optimize the communication cost together with finding a fusion configuration for minimizing storage. Finally, we minimize the data access times by minimizing disk-to-memory and memory-to-cache traffic and generate FORTRAN code. We have completed a first prototype of the TCE and are working on implementing the communication minimization and data access optimization algorithms. In future research, we will extend this approach to handle common subexpressions, symmetric matrices, and sparse matrices.
+
== Instructions / Details ==
 +
* [[user_details|For members of the lab]]
 +
* [[admin_details|For admins]]
  
=== Investigators ===
+
== Laboratory Members ==
  
* [http://www/.cse.ohio-state.edu/~saday P. Sadayappan]  (PI), Dept. of Computer and Info. Sci., Ohio State
+
=== Faculty ===
* [http://www.csc.lsu.edu/~gb Gerald Baumgartner], Dept. of Computer Sci., Louisiana State
+
* [http://www.csm.ornl.gov/~bernhold/ David E. Bernholdt], Oak Ridge National Lab
+
* [http://www.csm.ornl.gov/ccsg/html/staff/harrison.html Robert J. Harrison], Oak Ridge National Lab
+
* So Hirata, Pacific Northwest National Lab
+
* Marcel Nooijen, Dept. of Chemistry, University of Waterloo
+
* [http://www.chemistry.ohio-state.edu/cgi/brochure?Faculty=pitzer Russell M. Pitzer], Dept. of Chemistry, Ohio State
+
* [http://www.ee.lsu.edu/jxr/jxr.html J. Ramanujam], Dept. of Electrical and Computer Eng., Lousiana State
+
  
=== Publications ===
+
* [http://www.cse.ohio-state.edu/~saday/ Professor P. Sadayappan]
  
For an overview of the project, see our [http://www.csc.lsu.edu/~gb/TCE/Publications/SynthApproach2.pdf Proceedings of the IEEE paper]. A complete list of publications can be found [[TCE Publications|here]].
+
=== Researchers ===
 +
* Aravind Sukumaran Rajam
  
=== Software ===
+
=== PhD Students ===
 +
* [http://web.cse.ohio-state.edu/~rawatp/index.html Prashant Rawat]
 +
* [http://web.cse.ohio-state.edu/~baow/ Wenlei Bao]
 +
* [http://web.cse.ohio-state.edu/~nisa/ Israt Nisa]
 +
* Changwan Hong
 +
* Hyunjean Choi
 +
* Jinsung Kim
 +
* Rui Li
 +
* Süreyya Emre Kurt
 +
* Gordon E. Moon
  
There are several components available as part of the TCE software. Details can be found [[TCE Software|here]].
+
=== Master Students ===
 +
* Rakshith kunchum
 +
* Saurabh Singh
  
=== Funding ===
+
== Laboratory Alumni ==
 
+
* Samyam Rajbhandari, Intel
NSF  [http://www.itr.nsf.gov/ Information Technology Research Program], [https://www.fastlane.nsf.gov/servlet/showaward?award=0121676 Award #0121676], Sept. 2001 - Aug. 2006
+
* Martin Kong, Rice University
 +
* Venmugil Elango, NVIDIA
 +
* John Eisenlohr, Siemens
 +
* [https://sites.google.com/site/nsedaghati/ Naser Sedaghati], Imagination Technologies
 +
* [http://www.cse.ohio-state.edu/~niuq/ Qingpeng Niu], LinkedIn
 +
* [http://www.cse.ohio-state.edu/~fauzia/ Naznin Fauzia], Expedia
 +
* [http://www.cse.ohio-state.edu/~arafatm Md Humayun Arafat], Amazon
 +
* [http://www.mravishankar.com Mahesh Ravishankar], NVIDIA
 +
* [http://www.cse.ohio-state.edu/~laip/ Pai-Wei Lai], Yelp
 +
* [http://www.cse.ohio-state.edu/~ashari Arash Ashari], Facebook
 +
* [http://www.cse.ohio-state.edu/~tavarage/ Sanket Tavarageri], PlaceIQ
 +
* [http://kevinstock.org Kevin Stock], Two Sigma
 +
* Brian Larkins, Coastal Carolina University
 +
* James Dinan, Intel
 +
* Karthiyayini Chinnaswamy, Intel
 +
* Muthu Baskaran, Reservoir Labs
 +
* [http://www.cse.ohio-state.edu/~henretty Tom Henretty], Reservoir Labs
 +
* [http://jholewinski.org/ Justin Holewinski], NVIDIA
 +
* Giridhar S. Murthy, Apple
 +
* [http://hpc.pnl.gov/people/sriram/ Sriram Krishnamoorthy], Pacific Northwest National Laboratory
 +
* Nawab Ali, Intel
 +
* Arjun Singri, Amazon
 +
* [https://sites.google.com/site/qingda/ Qingda Lu], Intel
 +
* Gaurav Khanna, Intel
 +
* Rajkiran Panuganti, Microsoft
 +
* Alexander Sibiryakov, Microsoft
 +
* Sandhya Krishnan, Google
 +
* Uday Bondhugula, IISc
 +
* Nagavijayalakshmi Vydyanathan, Siemens
 +
* Aniruddha G. Shet, Intel
 +
* Gerald Sabin, RNET Technologies Inc.
 +
* Xiaoyang Gao, IBM
 +
* Bhargavi Rajaraman, Callfinity
 +
* Mohammad Kamrul Islam, LinkedIn
 +
* Alina Bibireata, GE Healthcare

Revision as of 21:12, 22 April 2018

Contents

High Performance Computing Research Projects

  • EAGER - Towards Automated Characterization of the Data-Movement Complexity of Large Scale Analytics Applications.
  • XPS: FULL - Collaborative Research: PARAGRAPH: Parallel, Scalable Graph Analytics.
  • PAdvisor - Padding advisor to avoid cache conflict misses.
  • Whole-Program Adaptive Error Detection and Mitigation.
  • Improving Vectorization.
  • Compiler/Runtime Support for Developing Scalable Parallel Multi-Scale Multi-Physics Applications.

  • Global Trees - Globally addressable distributed tree data structure library and runtime.
  • Pluto - An automatic parallelizer and locality optimizer for multicores.
  • Scioto - A scalable taskpools framework for global address space programming models.
  • Tensor Contraction Engine - An optimization system for quantum chemistry tensor equations.
  • UTS - The Unbalanced Tree Search benchmark is an irregular parallel benchmark.
  • PrimeTile - A parametric multi-level tiler for imperfect loop nests.
  • Hybrid MPI+UPC - Hybrid parallel programming with MPI and Unified Parallel C.
  • Polyhedral Compilation - Polyhedral Compilation projects.
  • DDGVec - A dynamic analysis of vectorization potential.
  • HOSTS - The High Order Stencil Transformation Suite: A framework for enhancing data reuse via associative reordering.
  • SDSL - Stencil Domain Specific Language: A language and compiler for developing performance portable stencil code.
  • IE - Inspector/Executor compiler : A source-to-source transformation to target distributed memory systems for partitionable irregular loop computations
  • SpMV on GPUs - Optimizing sparse matrix-vector multiplication on GPUs

Instructions / Details

Laboratory Members

Faculty

Researchers

  • Aravind Sukumaran Rajam

PhD Students

Master Students

  • Rakshith kunchum
  • Saurabh Singh

Laboratory Alumni

  • Samyam Rajbhandari, Intel
  • Martin Kong, Rice University
  • Venmugil Elango, NVIDIA
  • John Eisenlohr, Siemens
  • Naser Sedaghati, Imagination Technologies
  • Qingpeng Niu, LinkedIn
  • Naznin Fauzia, Expedia
  • Md Humayun Arafat, Amazon
  • Mahesh Ravishankar, NVIDIA
  • Pai-Wei Lai, Yelp
  • Arash Ashari, Facebook
  • Sanket Tavarageri, PlaceIQ
  • Kevin Stock, Two Sigma
  • Brian Larkins, Coastal Carolina University
  • James Dinan, Intel
  • Karthiyayini Chinnaswamy, Intel
  • Muthu Baskaran, Reservoir Labs
  • Tom Henretty, Reservoir Labs
  • Justin Holewinski, NVIDIA
  • Giridhar S. Murthy, Apple
  • Sriram Krishnamoorthy, Pacific Northwest National Laboratory
  • Nawab Ali, Intel
  • Arjun Singri, Amazon
  • Qingda Lu, Intel
  • Gaurav Khanna, Intel
  • Rajkiran Panuganti, Microsoft
  • Alexander Sibiryakov, Microsoft
  • Sandhya Krishnan, Google
  • Uday Bondhugula, IISc
  • Nagavijayalakshmi Vydyanathan, Siemens
  • Aniruddha G. Shet, Intel
  • Gerald Sabin, RNET Technologies Inc.
  • Xiaoyang Gao, IBM
  • Bhargavi Rajaraman, Callfinity
  • Mohammad Kamrul Islam, LinkedIn
  • Alina Bibireata, GE Healthcare
Personal tools