|
|
Wenli He |
|
||||
|
Ph.D. Candidate Department of Computer Science wenli-he@uiowa.edu |
Office: 101N MacLean Hall , Iowa City,IA 52242 USA Phone: (319)335-2839 |
|||||
|
|
||||||
|
Resume |
||||||
|
|
||||||
|
|
||||||
|
||||||
|
|
||||||
|
|
||||||
|
Design and
Implement an Effcient Parallel Algorithm for the Multi-Mode Resource
Constrained Project Scheduling Problem (MRCPSP) |
||||||
|
Department of Computer Science, The University of Iowa, USA, 2005 - now Project team member: Wenli He, Alberto Maria Segre |
||||||
 |
Multi-Mode Resource Constrained Project Scheduling Problem (MRCPSP) is a general model of the Project Scheduling Problem. This realistic model considers the project scheduling problem when resources such as machine, human resource, and budget are limited in availability with respect to the whole project duration or each time period during the project execution. Time-resource and resource-resource tradeoff is considered in form of multiple execution modes. Like most combinatorial optimization problems, the MRCPSP is NP-hard. Some problem instances with only 30 jobs are already beyond the computing power of a single processor. We designed a parallel algorithm which is both simple to implement and performance efficient. The network infrastructure our parallel application is implemented on is NICE (Network Infrastructure for Combinatorial Exploration). By using a Linux cluster consisting of only 10 quad-core 1.60GHZ Xeron processors, we've solved all the 640 instances of the nortorious 30-job benchmark problem set, denoted J30, generated decades ago by Kolisch et al. A research paper is in submission. |
|||||
|
|
||||||
|
|
||||||
|
Grid-enable and optimize large-scale parallel applications for Bayesian Geostatistical models on TeraGrid Academic Technologies - Research Services, the University of Iowa, USA, 2006 Project team member: Wenli He, Shaowen Wang, Jun Yan, Mary Kathryn Cowles, and Marc P. Armstrong |
||||||
  |
Bayesian geostatistical models enable statistical analysis and prediction based on data measured at irregularly-spaced geographic locations. The Markov chain Monte Carlo (MCMC) methods needed to fit these models require linear algebra operations that are computationally intensive when the number of measurement locations is large. Because an MCMC sampler typically must be run for thousands of iterations, each requiring numerous operations, the run-time for sequential Bayesian algorithms quickly becomes prohibitive. Even with parallel MCMC algorithms running on single clusters, run-times may be unacceptable, especially when large geographic datasets are analyzed. The TeraGrid provides an ideal platform to develop parallel MCMC algorithms for Bayesian geostatistical models by taking advantage of dynamically configurable Grid resources. We designed and implemented a parallel algorithm based on MPICH-G2 on the TeraGrid using PLAPACK for a class of Bayesian geostatistical models. The algorithm is designed to be scalable to the TeraGrid by exploiting its high-end network capabilities as well as using the parallel Cholesky decomposition function available from PLAPACK. The scalability of the algorithm is examined by varying physical block sizes and the number of processors used to compute large-scale linear algebra operations. Our algorithm achieves scalable speedups on single TeraGrid clusters. By running parallel Markov chains on multiple groups of processors, the level of scalability can be sustained when multiple TeraGrid clusters are used. - Receive the Finalist Award, CI-Impact Student Research Contest, TeraGrid 2006, Indianapolis, IN - Pictures are provided by Dr. Brian Smith to show a motivating problem: a lung cancer risk study based on residential radon concentrations. |
|||||
|
|
||||||
|
Develop an Apache Jetspeed portal for job submission and visualizing the parallel Bayesian Geostatistical application on TeraGrid Academic Technologies - Research Services, the University of Iowa, USA, 2006 |
||||||
  |
- An important contribution to the TeraGrid GIScience Gateway, an user-friendly portal for performing geographic information analysis using cyberinfrastructure capabilities.
|
|||||
|
|
||||||
|
Designed and developed a prototype Web Services Oriented Distributed WFMS, namely ISWF.
Institute of Software , Chinese
Academy of Sciences , China, 2001 - 2003 |
||||||
(ISWF System Architecture) |
The Web services technology provides the underpinning to a new business opportunity, i.e., the possibility of providing value-added Web services. Combining Web service with workflow technology can take advantage of Web services'characters, such as acting as "gray" boxes, depending highly on standards, loosely coupled interaction, and agility in integration, to upgrade the scalability and flexibility of the workflow system,and at the same time, build powerful Web service compositions with the type of coordination provided by workflow management. ISFlow is a workflow system focusing on building Web service compositions on top of heterogeneous and autonomous Web services.
|
|||||
|
|
||||||
|
|
||||||
|
Spring
2008
Fall
2007
Spring
2005
Fall 2004 |
22C:021:A01, 22C:021:A03 Computer Science II : Data Structures (Instructor) 22C:135 Theory of Computation (TA) 22C:166 Distributed Systems and Algorithms (TA) 22C:031 Algorithms (TA) 22C:031 Algorithms (TA) |
|||||
|
|
||||||
|
|
||||||
|
Wenli He, Alberto Maria Segre, "An Efficient Parallel Algorithm for the Multi-Mode Resource Constrained Project Scheduling Problem", in submission Wenli He, Shaowen Wang, Jun Yan, Mary Kathryn Cowles, and Marc P. Armstrong, "Using PLAPACK and MPICH-G2 to Grid-Enable Bayesian Geostatistical Models", (poster, presentation), TeraGrid'06 Conference, Indianapolis, IN, USA, Jun. 2006 Wenli He, Jun Wei, Chunyang Ye, " ISFlow-A Workflow System Supporting Web Services Compositions", in Proceedings of the International Workshop on Grid and Cooperative Computing(GCC2002), Hainan, China, December 2002, pp. 596-605 (link to www.once.com.cn) Wenli He, Jun Wei, "Web Service and Its Value in
EAI", Seventh Workshop for Graduate Students on Computer Science and
Technology, |
||||||
|
|
||||||
|
|
||||||
|
|
||||||
