Ruisheng Wang
SMART Interconnects Group
Department of EE-Systems
Los Angeles, CA 90089-2562
http://www-scf.usc.edu/~ruishenw
Cell: 213-***-****
Email: abp92a@r.postjobfree.com
Research Interests
Current Interests Onchip Network for Multi/Many-core Architecture, Onchip Resource (cache and memory
controller) Management in Chip Multiprocessor
Previous Interests High Performance Internet, High-Speed Router Design (Switch Architecture and Packet
Scheduler)
Education
Sep 2009-Current PhD student, Computer Engineering
University of Southern California, Los Angeles (GPA 3.835/4.0)
Sep 2006-Jul 2009 Master of Science, Computer Science & Technology
Tsinghua University, Beijing, China (GPA 88.78/100)
Sep 2002-Jul 2006 Bachelor of Science, Computer Science & Technology
North China University of Technology, Beijing, China (GPA 86/100)
Publications
1. Lizhong Chen, Ruisheng Wang, Timothy Mark Pinkston, Critical Bubble Scheme: An Efficient
Implementation of Globally-aware Network Flow Control, in Proceedings of the 25th IEEE International
Parallel & Distributed Processing Symposium (IPDPS 2011)
2. Yuho Jin, Ruisheng Wang, Woojin Choi, Timothy Mark Pinkston, Thread Criticality Support in On-Chip
Networks, in Proceeding of the 3rd International Workshop on Network On Chip Architectures (NoCArc
2010)
3. Ruisheng Wang, Youjian Zhao, Hongtao Guan and Guanghui Yang, HOBRP: A Hardware Optimized
Packet Scheduler that Provides T unable End-to-end Delay Bound, in Proceedings of 17th IEEE International
Workshop on Quality of Service (IWQoS 2009)
4. Ruisheng Wang and Youjian Zhao, Achieving 100% Throughput in a Two-stage Multicast Switch, in
Proceedings of the 23rd International Conference on information Networking (ICOIN2009)
5. Zhenhua Liu, Xiaoping Zhang, Youjian Zhao, Ruisheng Wang, A Performance Analysis Framework for
Routing Lookup in Scalable Routers, in Proceedings of the 23rd International Conference on information
Networking (ICOIN2009)
Projects
May2011
Coordinated Management for Cache Capacity and Memory Bandwidth in CMP --advisor
-Current
Prof. Timothy M. Pinkston
As new multi-threaded usage models such as virtualization and consolidation take advantage of multiple cores in
CMP architectures, the impact of shared resource contention between VMs and user-level applications introduces
Quality of Service(QoS) concerns and challenges. Our goal of this research is to manage multiple shared CMP
resources in a coordinated fashion to optimize system level performance objectives.
May 2010-May 2011
Thread Criticality in On-Chip Networks -advisor Dr. Yuho Jin
In multi-threaded application, the critical thread may delay all the other threads due to the synchronizations. We
proposed to dynamically build fast bypass path for critical threads, to boost overall system performance.
2011 Spring
Optimizing LU Decomposition in GPU (course project)
We implemented and analyzed the LU decomposition algorithm in Nvida GPU platform.
2010 fall
Access Patten Aware Cache Placement in a Hybrid Shared/Private NUCA (course project)
I proposed an intelligent replication policy based on a runtime cache block classification mechanism in a hybrid
shared/private NUCA. Cache blocks are classified into three groups: private, infrequently updated shared and
frequently updated shared. The first two groups are organized into private caches to improve performance while
the last one is organized into shared cache to reduce coherence overhead. The proposed mechanism was evaluated
with the trace of multi-threaded applications in PARSEC benchmark suite.
2010 Spring
Evaluation of DRAM power management algorithms in CMP environment (course project)
The efficiency of most DRAM power management algorithms is highly depended on its accuracy of memory
access prediction. Memory accesses in CMP are lack of locality, which make them hard to be predicted. The goal
of this project is to analyze the efficiency of three DRAM power management algorithms in the context of Chip
Multiprocessor. We re-evaluated three power management algorithms (LRU-Greedy, LRU-Smooth, LRU-
Ordered) in a 16 core CMP environment with a full system simulator SIMICS with GEMS.
2010 Spring
Comparison of Virtual and Physical Express Topologies in GARNET (course project)
We compared Express Virtual Channel (virtual bypasses) and Flatten Butterfly (physical bypasses) under various
synthetic and realistic (PARSEC benchmarks in SIMICS) traffic in GARNET.
Apr 2008-
FPGA-based switching system -Advisor: Prof. Youjian Zhao
Jul 2009
Master Graduation Project
I designed and implemented a 4 4 buffered crossbar in a FPGA chip. The crossbar has four GbE ports with
round-robin scheduler for both input and output. The design was synthesized by Quartus II and verified by
ModelSim.
Oct 2007-
CNGI backbone network monitoring and performance analyzing system
Nov 2008
(Part of China Next Generation Internet (CNGI) Project)
-Advisor: Prof. Youjian Zhao
The job of our hardware group in this project is to design a network device, which can distribute the network
traffic from an OC192 link to eight Gigabit Ethernet ports according to pre-defined rules (e.g. forward a particular
flow to a particular port). As a core member of this project, I wrote the drivers for the hardware (e.g., configuring
framers (PM5390/IXF1104) and custom FPGA, implementing CPU interrupts (MPC750)) and command-line
user interface for the system. I also implemented a packet forwarding function in a FPGA, and performed system
debugging in latter stage of the project.
Skills
Projects in C/C++ with MPI and OpenMP CUDA, Java (Jxta, JMP), VHDL,
Familiar with Python, Visual Basic, SHELL, TCL, ASP SQL, HTML,
Tools Simics/GEMS, Gem5, Tornado (VxWorks), Visual Studio, Eclipse, NS2(network
simulator), Matlab, Cadence, HSpice, QuartusII, ModelSim
Coursework (all taken from USC)
Computer Systems Architecture (EE557) Special Topics in Energy Efficiency and Reliability in ICT (EE599)
Parallel Programming (CS503) Broadband Network Architecture (EE555)
Interconnection Networks (CS659) Advanced Operating Systems (CS555)
VLSI System Design (EE477L/EE577a) Parallel and Distributed Computing (EE657)
Advanced Topics in Micro-architecture (EE653) Analysis of Algorithms (CS570)