General Information
|
Note: I've graduated from CS@UMN at Sept, 2011
Education
|  |
Awards & Honors
- Student Travel Grants for the 19th Intl. Conf. on Parallel Architectures and Compilation Techniques, Sept 2010
- Nominee for IBM PhD Fellowship, Dept. of Computer Science & Engineering, UMN, Oct 2009
- Student Travel Grants for the 18th Intl. Conf. on Parallel Architectures and Compilation Techniques, Sept 2009
- Student Travel Grants for the 36th International Symposium on Computer Architecture, Jun 2009
- Best Paper Award of the 26th IEEE International Conference on Computer Design (ICCD), Oct 2008
- Nominee for IBM PhD Fellowship, Dept. of Computer Science & Engineering, UMN, Oct 2008
- Nominee for Graduate Fellowship upon Admission at University of Minnesota - Twin Cities, Jan 2007
Research Experience
- Advisor: Antonia Zhai
- Area of Focus: Computer Architecture, Multicore Processors, Speculative Parallelization
Research Projects:
- Background: Thread-Level Speculation (TLS) and Speculative Parallelism
TLS is a technique for speeding up single-threaded programs on multi-threaded architectures. Those programs are often difficult to parallelize due to possible data dependences that are hard to detect during compile time. The TLS approach allows the compiler to speculatively generate parallel code regardless of the possible dependences and employs hardware components to track and recover dependence violation at runtime. This type of parallelism is regarded as speculative parallelism. - Topic I: Improving Performance for Thread-Level Speculation (TLS)
TLS is a technique for speeding up single-threaded programs on multi-threaded architectures. The challenge is to select the appropriate regions for TLS. Prior approaches rely on compiler analysis with program profiling. But they suffer from inaccuracy in modeling many runtime aspects. My approach is to monitor runtime execution and determine the region selection. - Topic II: Improving Energy Efficient for Speculative Threads
TLS improves performance at the cost of energy efficiency. I propose a heterogeneous multicore design for TLS, and dynamic allocation heuristics that match threads with the most energy-efficient configuration.
Publications:
- Y. Luo and A. Zhai. "Dynamically dispatching speculative threads to improve sequential execution." ACM Transactions on Architecture and Code Optimization (TACO), Volume 9 Issue 3, September 2012
Acceptance Rate: 26% - Y. Luo. Exploiting parallelism in multicore processors through dynamic optimizations. ProQuest, UMI Dissertation Publishing, 2012 (University of Minnesota PhD Thesis, ISBN: 978-1249905264)
- Y. Luo, V. Packirisamy, W. Hsu and A. Zhai. "Energy Efficient Speculative Threads: Dynamic Thread Allocation in Same-ISA Heterogeneous Multicore Systems." The 19th International Conference on Parallel Architectures and Compilation Techniques (PACT), Vienna, Austria, September 2010.
Acceptance Rate: (46/266) 17% - X. Zhuang, A. Eichenberger, Y. Luo, K. O'Brien and K. O'Brien. "Exploiting Parallelism with Dependence-Aware Scheduling." The 18th International Conference on Parallel Architectures and Compilation Techniques (PACT), Raleigh, North Carolina, USA, September 2009.
Acceptance Rate: 35/188 (18%) - Y. Luo, V. Packirisamy, N. Mungre, A. Tarkas, W. Hsu and A. Zhai. "Dynamic Performance Tuning for Speculative Threads." The 36th International Symposium on Computer Architecture (ISCA), Austin, Texas, USA, June 2009.
Acceptance Rate: 43/210 (20%) - V. Packirisamy, Y. Luo, W. Hung, A. Zhai, and P. Yew. "Efficiency of Thread-Level Speculation in SMT and CMP Architectures - Performance, Power and Thermal Perspective." The 26th IEEE International Conference on Computer Design (ICCD), Lake Tahoe, California, USA, October 2008.
Acceptance Rate: 106/306 (34%) (Best Paper Award)
Patent(s):
- A. Eichenberger, Y. Luo, K. O'Brien and X. Zhuang
"Methods for Parallelization of Irregular Reductions via Parallel Building and Exploitation of Conflict-Free Units of Work at Runtime."
United States Patent Application: 2011-0088020 A1
Assignee: IBM Corporation, Armonk, NY
Internship(s)
- SeaMicro, Inc.
Software Engineering Fabric routing: algorithm, system design and implementation, and performance tuning
July - August 2011 - IBM T.J. Watson Research Lab Yorktown Site.
Advanced Compiler Technology Team, under Kathryn O'Brien
Project: Dependence-Aware Scheduling Using Assistant Threads
May - August 2008
Scholarly Addresses
Sept 2010: Vienna, Austria
- Energy Efficient Speculative Threads: Dynamic Thread Allocation in Same-ISA Heterogeneous Multicore Systems
- The 19th International Conference on Parallel Architectures and Compilation Techniques
Oct 2009: Minneapolis, Minnesota
- Dynamic Performance Tuning for Speculative Threads
- The 7th Biennial Open House, Department of Computer Science and Engineering, UMN
June 2009: Austin, Texas
- Dynamic Performance Tuning for Speculative Threads.
- The 36th International Symposium on Computer Architecture
Aug 2008: Yorktown Heights, New York
- Dependence-Aware Scheduling Using Assistant Threads
- IBM Summer Intern Poster Session, IBM T.J. Watson Yorktown Site
Feb 2008: Salt Lake City, Utah
- Compiler Optimization for Parallelizing General-Purpose Applications under Thread-Level Speculation
- The 13th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
Teaching Experience
Guest Lecturer (Spring 2011, Bethel University)
- COS105 Computer Science I, and COS216 Data Structures and Objects
- Instructor: Nathan Gossett
- Duties: delivering three distinctive lectures and getting observed and mentored by the instructor, part of the University of Minnesota Preparing Future Faculty (PFF) program.
Teaching Assistant (Spring 2011)
- CSci 4041: Algorithms and Data Structures
- Main Instructor: Carl Sturtivant
- Duties: leading weekly recitation sessions, grading homework/exams, maintaining grades, holding office hours
- Midterm Student Evaluation: 4.27/5, 18 surveys
Item Score Item Score 1. Knowledgeable about the material: 4.56/5 2. Speak audibly and clearly: 4.33/5 3. Has good rapport with students: 4.28/5 4. Invites questions and discussions: 4.28/5 5. Provides clear answers to questions: 4.17/5 6. Summarized or emphasizes key points: 4.17/5 7. Makes good use of examples: 4.39/5 8. Encourages students to think: 4.33/5 9. Provides helpful feedback: 4.22/5 10. The discussion section useful: 4.00/5 - End-semester Student Evaluation: 5.17/6, 14 surveys
Item Score Item Score 1. Well-prepared for class: 5.21/6 2. Presented the subject clearly: 5.00/6 3. Provided feedback for students' performance: 5.00/6 4. Treated students with respect: 5.57/6 5. Students have a deeper understanding: 5.14/6 6. Students' interest stimulated: 5.07/6
Guest Lecturer (Spring 2011)
- CSci 8205: Parallel Machine Organization
- Main Instructor: Prof Antonia Zhai
- Duties: Delivering lectures on advanced research topic sessions: Speculative and Transactional Execution
Guest Lecturer (Fall 2010)
- CSci 5204: Advanced Computer Architecture
- Main Instructor: Prof Antonia Zhai
- Duties: Delivering lectures on research topic sessions: Simultaneous Multithreading, Processor Energy Efficiency
Teaching Assistant (Fall 2008)
- CSci 2021: Machine Architecture and Organization
- Instructor: Prof Antonia Zhai
- Duties: grading homework/exams, setting up programming assignments, leading recitation sessions, maintaining grades, holding office hours
Teaching Assistant (Fall 2007)
- CSci 4011: Formal Languages and Automata Theory
- Instructor: Prof Gopalan Nadathur
- Duties: leading recitation sessions, grading homework/exams, maintaining grades, holding office hours
- Student Overall Rating: 4.15/5, 20 surveys
Item Score Item Score 1. Approachable in office hour: 4.67/5 2. Good grasp of material: 4.50/5 3. Helpful in office hour: 4.29/5 4. Good communication skills: 3.60/5 5. Well-prepared & organized: 4.45/5 6. Present material clearly: 4.00/5 7. Grade consistently and clearly: 4.10/5
Course Works
- Overall GPA: 3.919/4
Spring 2011
- GRAD 8102: Practicum For Future Faculty
Instructor: Christina Petersen
Final Grade: S
Spring 2010
- GRAD 8101: Teaching in Higher Educations
Instructor: Ilene Alexander and Esam El-Fakahany
Final Grade: S
Fall 2009
- EE 5301: VLSI Design Automation I
Instructor: Sachin Sapatnekar
Final Grade: S
Spring 2009
- MATH 5707: Graph Theory and Non-enumerative Combinatorics
Instructor: Drew Armstrong
Final Grade: A- - CSCI 5161: Introduction to Compilers
Instructor: Wei-Chung Hsu
Final Grade: A
Fall 2008
- EE 5371: Computer Systems Performance Measurement and Evaluation
Instructor: David J. Lilja
Final Grade: A - CSci 5511: Artificial Intelligence I
Instructor: Maria L. Gini
Final Grade: A - CSci 5801: Software Engineering I
Instructor: Phillip Barry
Final Grade: A-
Spring 2008
- CSci 8205: Parallel Computer Organization
Instructor: Antonia Zhai
Final Grade: A - CSci 5451: Introduction to Parallel Computing
Instructor: Karypis, George
Final Grade: A - CSci 8980: Special Advanced Topics in Computer Science - Recent Advanced in Computer Architectures
Instructor: Pen-Chung Yew
Final Grade: A- - CSci 8002: Introduction to Research in Computer Science II
Instructor: Joseph A. Konstan
Final Grade: A - CSci 8970: CSCI Colloquium
Instructor: variety
Final Grade: S