MOLAR Homepage

Homepage | People | Projects | Publications

MOLAR: Modular Linux and Adaptive Runtime Support for
High-end Computing Operating and Runtime Systems^*

MOLAR is a multi-institution research effort that concentrates on adaptive, reliable, and efficient operating and runtime system solutions for ultra-scale high-end scientific computing on the next generation of supercomputers. This research addresses the challenges outlined by the FAST-OS - forum to address scalable technology for runtime and operating systems - and HECRTF - high-end computing revitalization task force - activities by providing a modular Linux and adaptable runtime support for high-end computing operating and runtime systems.

Current operating systems and runtime systems (OS/Rs) are not able to meet the various requirements to run large applications efficiently on future ultra-scale computers. This effort's approach is to develop and experiment with various variations of Linux operating systems ranging from light-weight -to- custom -to- full and then to provide a crosscut of functionality through the high-end computing (HEC) OS research map as a proof-of-concept implementation, so that other researchers and developers in operating systems, tools, middleware, and applications may directly leverage our work.

Building on the current open-source operating system, Linux, we target HEC applications for the next generation of supercomputers. Undoubtedly, these HEC OS/Rs must scale to the levels predicted by hardware architects for both shared memory and distributed memory platforms; furthermore, they must enable applications to operate efficiently and reliably on any of these architectures as transparently as possible. As described in recent reports by FAST-OS, HECRTF and DoE Scales activities, system software is a key challenge in exploiting the promise of extreme-scale scientific computing. Conceptually, the MOLAR research has the following goals to address these issues.

Create a modular and configurable Linux system that allows customized changes based on the requirements of the applications, runtime systems, and cluster management software.
Build runtime systems that leverage the OS modularity and configurability to improve efficiency, reliability, scalability, ease-of-use, and provide support to legacy and promising programming models.
Advance computer reliability, availability and serviceability (RAS) management systems to work cooperatively with the OS/R to identify and preemptively resolve system issues.
Explore the use of advanced monitoring and adaptation to improve application performance and predictability of system interruptions.

Furthermore, through our relationship with industry partners, such as Cray Inc., we investigate the support of the MOLAR OS/R directly in more advanced ultra-scale architectures. Finally, it is our intent that this research effort contributes back into the OS community through cooperation in both providing our work in an open source format to the community as well as through our effort to gain acceptance of developed research codes in the general community OS efforts.

^*This research is sponsored by the Office of Advanced Scientific Computing Research; U.S. Department of Energy. The work is performed jointly at Oak Ridge National Laboratory, which is managed by UT-Battelle, LLC under Contract No. De-AC05-00OR22725, Louisiana Tech University, Ohio State University and North Carolina State University in collaboration with University of Reading and Cray Inc..