hiperspace lab
Computer and Information Sciences
University of Delaware



This is on old version of our web site. See our new web site: Software Analysis and Compilation Research Lab

Welcome to the home page of the HiperSpace Research Group! HiperSpace consists of a group of faculty, graduate students, and undergraduates in the Department of Computer and Information Sciences at the University of Delaware. Our research addresses various aspects of program analysis and transformation for code optimization and software tools, targeted toward software engineering and high performance computer architectures. Current research projects focus on mobile code validation through static program analysis, analysis and testing of web-based software systems, online impact analysis for software maintenance, testing program-based security mechanisms, techniques and tools for aspect mining, and developing an integrated approach to improve communication performance in high performance clusters.

We are constantly changing our membership, and often have opportunities for new members to join our research group as a graduate student doing Ph.D. research, as a master's student interested in doing a master's thesis, or as an undergraduate student interested in summer undergraduate research or an honors thesis. If you are interested in discussing the possibility of joining our research group, send email to Lori Pollock to set up an appointment.

Undergraduate Research Opportunities


Professors Lori Pollock and Vijay Shanker are featured in an article on UDaily talking about the NSF grant that they have received. This grant is for their seminal work in introducing Natural Language Processing techniques to the field of program analysis, thereby helping in a variety of program comprehension and maintenance tasks.

Here is a link to the article (our archived version) UD researchers mine words to help programmers


HiperSpace students are currently working on projects on mobile code security, testing of web applications, aspect-oriented program analysis and mining, optimizations for cluster parallel programming, and exploiting dynamic compilers for dynamic program analysis.

We have a research project overview in PowerPoint and more complete information on the projects here.



Complete List of HiperSpace Publications

Publications of Current Projects:

If the above links are not working or you're not finding what you're looking for, please refer to our traditional publications page.



SIGSYS Current schedule

SIGSYS is a weekly seminar discussing recent papers of relevance to issues in systems, high-performance computing, program analysis, and compilation techniques.

Topics: effective program compilation, programming languages and paradigms, software tools, and analysis and optimization techniques for high performance architectures and computing systems, including high performance microprocessors, vector, parallel and distributed computing systems, and networks of workstations.

Speakers include leading researchers from academia, industry, and within the department.

Sign up for CISC 890-016 CLQ:SIG:High Performance Computing

For more information about SIGHPC contact Dr. Pollock

SIGHPC Fall 2005 schedule
SIGPACT Fall 2003 schedule
SIGPACT Spring 2003 schedule
SIGPACT Fall 2002 schedule
SIGPACT Spring 2002 schedule
SIGPACT Fall 2001 schedule


  Graduate Coursework Most Relevant to the Research
(Courses in ECE may also apply.)
  • CISC 662 Computer Systems Architecture
  • CISC 672 Advanced Compiler Construction
  • CISC 872 Advanced Program Analysis and Transformations
  • CISC 873 Compiling for Advanced Architectures
  • CISC 874 Parallel Programming - Overview of high performance computing architectures, parallel program performance measurement, different paradigms for achieving parallelism including automatic parallelization, message passing, shared memory, data parallel programming with HPF, task and loop scheduling, debugging parallel programs, issues in getting good performance on parallel architectures. PREREQ: CISC 662 is recommended.
  • CISC 879 Software Tools and Environments (Spring 2007)
  • CISC 879 Software Testing and Maintenance (Spring 2004)


  • NSF - The National Science Foundation (Grants 0702401, 0509170, and 0720712)
  • CTA - Army Research Lab Collaborative Technology Alliance
  • ARL - The Army Research Laboratory


  We are not far off of Interstate-95 in Delaware.
    From the North
    • Take exit #1B: 896 North (following the signs for the University of Delaware)
    • Follow 896 North, which becomes South College Avenue if you go straight (896 will turn to the left)
    • Keep going straight on South College to the University's Main Campus (approx. 2 miles)
    • Smith Hall is located on your left on the corner of South College Avenue and Amstel Avenue (right next to the pedestrian over-pass)
    • Visitor parking is available on the left on South College Avenue just before Smith Hall (Lot 41 in parking map, look for the University Visitor's Center) or in the parking garage located on Main Street (Lot 17 in parking map)
    From the South
    • Take exit 109 A-B in Maryland to avoid the toll at the DE line. Take the far ramp toward Newark, which will put you on Elkton-Newark Road.
    • You will be on Elkton Road for 3-4 miles until it merges into one lane and curves right, where it changes into Delaware Avenue.
    • At the light, turn RIGHT onto South College Avenue. Smith Hall is the second building on your right
    • Visitor parking is available
      • in a lot on the right on South College Avenue (Lot 41 in parking map, look for the University Visitor's Center) and
      • in a parking garage on Amstel Avenue. Make a RIGHT at the light onto Amstel Avenue and a LEFT at the next stop sign. Parking garage is on your RIGHT (Lot 14C in parking map).

Bring your parking receipt to the CIS Office for validation.

For more info and maps, check here.


  Affiliated labs, professional organizations, libraries, conference info, compiler infrastructures, groups, etc . . . .

more here...


  A little bit of history about HiperSpace . . .

All Archived Publications For Search Engines