The SciTech group does research and development on software systems to help scientists manage large-scale computations. We work with scientists in domains ranging from genomics and proteomics to seismology and gravitational wave physics. We help scientists deploy their computations on some of the largest and fastest computing systems in the world. Our main project is the Pegasus Workflow Management system, which is used to orchestrate complex, large-scale data processing and computation pipelines. Pegasus compiles abstract, high-level workflow descriptions into efficient executable workflows that can be deployed on diverse cyberinfrastructure. In this process it expands and optimizes the workflow, adding data management tasks and performing task clustering, cleanup and other optimizations.
The SciTech group is part of the USC Information Sciences Institute (ISI). A unit of the University of Southern California’s highly ranked Viterbi School of Engineering, ISI is one of the nation’s largest, most successful university-affiliated computer research institutes. ISI researchers bridge the gap between theoretical basic research and product-oriented research and development. Its diverse expertise ranges from core engineering and computer science discovery to design, modeling and implementation of innovative prototypes and devices. A pacesetter for more than 40 years, ISI helped conceive, design and implement the Internet, including communications protocols that remain fundamental to Net operations such as TCP/IP, ICMP, DNS and SMTP. Today ISI’s research initiatives span a broad range of fields including: intelligent systems such as natural language processing, machine translation and information integration; informatics that include computer networks, medical informatics and decision systems; computational systems such as supercomputing approaches and biomimetics; and advanced electronics for space and other purposes.
ISI is located in beautiful, sunny Marina Del Rey, California. Our offices overlook and are just minutes from the beach.
Current open positions:
Postdoctoral Scholar – Research Associate
The ideal candidate would have strong system programming skills with experience in distributed computing and scientific computing.
- Develop new tools and leverage existing tools to profile scientific applications for I/O and performance characteristics.
- Development of methodologies for collection and dissemination of multi-sourced performance and application data.
- Development of ML techniques for workflow performance analysis and infrastructure troubleshooting
- Takes lead in publishing the research in related conferences and journals.