Meng, Jiayuan

I am currently an Argonne Scholar at Argonne National Laboratory's Leadership Computing Facility Division. I received my Ph.D. from Department of Computer Science in University of Virginia in Aug. 2010. My research focus is computer architecture and high performance computing. I am also interested in graphics, artificial intelligence and machine learning.               Contact me

 

 

Projects:

Fractal: Streaming CMP Architecture Simulator

MV5: A Reconfigurable Simulator for Heterogeneous Multicore Architectures

Advised by Prof. Kevin Skadron.

[a patch for M5 that supports multithreading in SE mode, directory coherence, and inteconnects (2D meshes)]

Characterizing on Parallel Applications:

Advised by Dr. Dee Weikle

Shaping with Patterns:

Course project on Jason Lawrence's class Data Driven Models in Computer Graphics in Fall, 2006.

[report] [src code(not documented yet)]

Non-Parametric Texture Synthesize:

Implemented during Jason Lawrence's class Data Driven Models in Computer Graphics in Fall, 2006.

Using Markov Random Field possibility theory to grow a new texture from a 3 by 3 "seed" from the original texture. The new texture will resemble the original texture.

[Efros and Leung et.al]

Image Analogies:

Implemented during the Jason Lawrence's class Data Driven Models in Computer Graphics in Fall, 2006.

[Hertzmann et.al]

 Poisson Editting 

Implemented during the Jason Lawrence's class Data Driven Models in Computer Graphics in Fall, 2006.

[perez et.al 2003]

Key Feature Matching

Implemented during the Jason Lawrence's class Data Driven Models in Computer Graphics in Fall, 2006.

Using SIFT to find the scale and rotation invariant key points.

[Lowe et.al 2004]

AmberBBV: [AmberBBV main page]

Supervised by Prof. Kevin Skadron and Dr. Dee Weikle. Here I am building a tool to capture the basic block vectors at run time according to the traces of PowerPC applications captured by amber. This is an amber plug-in using amber externel library. 

Micro-architecture independent characterizing on Game applications:

Supervised by Prof. Kevin Skadron and Dr. Dee Weikle. Using micro-architecture independent characteristics to evaluate games such as WarCraft 3, Doom 2 and a ray tracer PBRT.  Published as TR CS-2007-04

Fluid Simulation with a Multigrid PDE Solver on GPU

 Mark Harris has wrote a Fluid Simulation on a 2D grid. This small project tries to extend it by using multi-grid PDE solver on GPU.

download code (code may not work if GPU don't have enough support)

video: [no viscosity] [low viscosity] [high viscosity] [vector fields]

GPU implementation of Light-Field Photography:

Supervised by Prof. Kevin Skadron, Dr. Dee Weikle, and Prof. Greg Humphreys. Ren Ng and Mark Levoy have published the idea of the light field camera using microlens arrays. This implementation trys to accelerate the computation by using SSE2, graphics hardware, and Cell Broad Band Engine(PlayStation3 processor). It can run 35 fps for a 4096*4096 lightfield image, user can browse different depth at real time.

[tech report]  code for: [CBE] [GPU][SSE]

Game Engine:

Course project in the coarse Real Timer Rendering on Spring, 2006  taught by Prof. David Luebke. This is a joint project together with Jiajun Zhu and Kevin Dale.We were focusing on the rendering engine.

(download video high|low)

Virtual Creature: This is a course project together with Jiajun Zhu for Intro Graphics in Fall, 2005 taught by Prof. Greg Humphreys. The key technique used here is artificial neural network and genetic programming.  It tries to reproduce Karl Sims' work about "Evolved Virtual Creatures". Can't believe it that the creature got evolved just 2 days before the Demo! Life is a miracle!

[html](Youtube)(download video high | low , video for brain structure: brain )

High-Resolution Display Wall

This is my senior thesis project under the supervision of Prof. Jiaoying Shi and Prof. Hai Lin in Zhejiang University. Using 5*3 distributed PCs each with a projector, display a large high-resolution image on the screen seamlessly.

This work has been published. See:

[Download pdf]

Education

 

Workshop Organization

 

Publications

2013 Characterization and Understanding Machine-Specific Interconnects Vitali Morozov, Jiayuan Meng, Venkatram Vishwanath, Kalyan Kumaran, Michael E. Papka. In the Proceedings of the 12th International Conference on Parallel Computing Technologies, PaCT 2013
Online Performance Projection for Clusters with Heterogeneous GPUs. L. S. Panwar, A. M. Aji, J. Meng, P. Balaji, and W. Feng. In the Proceedings of the IEEE International Conference on Parallel and Distributed Systems (ICPADS). 2013.
Early Experience on the Blue Gene/Q Supercomputing System V. Morozov, V. Vishwanath, K. Kumaran, J. Meng, M. Papka. In Proceedings of the 27th IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2013.
A Multiple SIMD, Multiple Data (MSMD) Architecture: Parallel Execution of Dynamic and Static SIMD Fragments. Y. Wang, S. Chen, J. Wan, J. Meng, K. Zhang, W. Liu, and X. Ning. The 19th IEEE International Symposium on High Performance Computer Architecture (HPCA), 2013
SESH framework: A Space Exploration Framework for GPU Application and Hardware Codesign. J. Lee, J. Meng, H. Kim. In the 4th International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS13), held in conjunction with the Supercomputing Conference (SC). 2013
Improving GPU Performance Prediction with Data Transfer Modeling. M. Boyer, J. Meng, K. Kumaran, In the 3rd International Workshop on Accelerators and Hybrid Exascale Systems (AsHES), 2013
2012 Dataflow-Driven GPU Performance Projection for Multi-Kernel Transformations. J.Meng, V. Morozov, V. Vishwanath, and K. Kumaran, In ACM/IEEE Supercomputing Conference, Nov 2012 Slides
  ALCF MPI Benchmarks: Understanding Machine-Specific Communication Behavior. V. Morozov, J. Meng, V. Vishwanath, J. Hammond, K. Kumaran, and M. Papka. In the Proceedings of the P2S2 Workshop, Sept. 2012
  Robust SIMD: Dynamically Adapted SIMD Width and Multi-Threading Depth J. Meng, J. Sheaffer, and K. Skadron, To appear in the 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2012.
2011 GROPHECY: Projecting GPU Performance from CPU Code Skeletons J. Meng, V. Morozov, K. Kumaran, V. Vishwanath, and T. Uram In Proceedings of the ACM/IEEE Supercomputing Conference,  Nov 2011 Slides
  A reconfigurable simulator for large-scale heterogeneous multicore architectures J. Meng, K. Skadron. In Proceedings of the IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), 2011
2010 Dynamic Warp Subdivision for Integrated Branch and Memory Divergence Tolerance J. Meng, D. Tarjan, and K. Skadron  In Proceedings of the 37th ACM/IEEE International Symposium on Computer Architecture, June 2010
 

Exploiting the Forgiving Nature of Applications for Scalable Parallel Execution J. Meng, A. Raghunathan, S. Chakradhar, and S. Byna, In Proceedings of the 24rd International Parallel and Distributed Processing Symposium, 2010

  Exploiting Inter-thread Temporal Locality for Chip Multithreading J. Meng, J. Sheaffer, and K. Skadron. In Proceedings of the 24rd International Parallel and Distributed Processing Symposium, 2010
2009 Avoiding Cache Thrashing due to Private Data Placement in Last-level Cache for Manycore Scaling J. Meng and K. Skadron In IEEE International Conference on Computer Design,  Oct 2009
  Increasing Memory Latency Tolerance for SIMD Cores D. Tarjan, J. Meng and K. Skadron In ACM/IEEE Supercomputing Conference,  Nov 2009 (Best Student Paper Award)
  Performance Modeling and Automatic Ghost Zone Optimization for Iterative Stencil Loops on the Tesla Architecture  J. Meng and K. Skadron.  In the 23rd International Conference on Supercomputing, June 2009
  Leveraging Memory Level Parallelism Using Dynamic Warp Subdivision J. Meng, D. Tarjan and K. Skadron.  UVA Tech Report CS-2009-02.
  Best-Effort Parallel Execution Framework for Recognition and Mining Applications J. Meng, S. Chakradhar, and A. Raghunathan. In the proceedings of the 23rd International Parallel and Distributed Processing Symposium, May 2009
  Rodinia: A Benchmark Suite for Heterogeneous Computing Shuai Che, Michael Boyer, Jiayuan Meng, David Tarjan, Sang-Ha Lee, Jeremy Sheaffer, and Kevin Skadron. IISWC, Oct. 2009
2008 Programming with Relaxed Streams. J. Meng, S. R. Tarapore, S. Che, J. Huang, J. W. Sheaffer, and K. Skadron. UVA Tech. Report CS-2007-17
2007 A Performance Study of General Purpose Applications on Graphics ProcessorsS. Che, J. Meng, J. W. Sheaffer, and K. Skadron. First Workshop on General Purpose Processing on Graphics Processing Units, Northeastern University, Oct. 2007
  Paraweaver: Performance Evaluation on Programming Models for Fine Grained Threads  J. Meng, D. A. B. Weikle, K. Hazelwood. UVA Tech. Report CS-2007-09
  Comparing Doom 3, WarCraft III, PBRT and MESA Using Micro-Architectually Independent Characteristics J. Meng, H. Cook, K. Skadron, D. A. B. Weikle. Technical Report CS-2007-04
 

An Approach on Hardware Design for Computational Photography Applications based on Light Field Refocusing Algorithm  J. Meng,, D. A. B. Weikle, G. Humphreys, K. Skadron, UVA Tech Report CS-2007-15

2006 High-Resolution Image Viewing on Projection-based Tiled Display Wall J. Meng, H. Lin, and J. Shi. Proceedings of SPIE on Electronic Imaging 2006
   

Awards

2010: UVA Award for Excellence in Scholarship in the Sciences & Engineering

2009 - 2010: NVIDIA PhD Fellowship

2002 - Now:  Melton Fellow

Locations of visitors to this page