Past CLaSS Group Projects

ASIM: An Agent-based Integrated Model for Complex Networks Complex networks comprised of large-scale distributed heterogeneous systems are everywhere, for example, in brains, networks of social and economic interactions, and the Internet. Often these systems have surprising common properties, such as power-law degree distributions, which occur in networks as different as metabolic networks and co-authorship patterns in scientific publications. Many mathematical and computational models have been developed over the past decade that attempt to explain emergent properties at a macro-level. However, understanding more detailed structure and behavior of complex networks has required the incorporation of more domain-specific features, for example, incorporating spatial distributions and business relationships into models of the Internet. This project focuses on the Internet, and Internet-like systems.

Berkeley Lab Checkpoint/Restart (BLCR) for LINUX This work focuses on checkpointing parallel applications that communicate through MPI, and on compatibility with the software suite produced by the SciDAC Scalable Systems Software ISIC.

Berkeley Container Library The Berkeley Container Library (BCL) is a set of generic, cross-platform, high-performance data structures for irregular applications, including queues, hash tables, Bloom filters and more. BCL is written in C++ using an internal DSL called the BCL Core that provides one-sided communication primitives such as remote get and remote put operations. The BCL Core has backends for MPI, OpenSHMEM, GASNet-EX, and UPC++, allowing BCL data structures to be used natively in programs written using any of these programming environments.

BQSKit The Berkeley Quantum Synthesis Toolkit (BQSKit) is a superoptimizing quantum compiler and research vehicle that combines ideas from several projects at LBNL into an easily accessible and quickly extensible software suite.

CORVETTE: Correctness Verification and Testing of Parallel Programs The goal of this project is to provide tools to assess the correctness of parallel programs written using hybrid parallelism. There is a dire lack of both theoretical and engineering know-how in the area of finding bugs in hybrid or large scale parallel programs, which our research aims to change. As intranode programming is likely to be the most disrupted by the transition to Exascale, we will emphasize support for a large spectrum of programming and execution models such as dynamic tasking,…

DEGAS: Dynamic Exascale Global Address Space Programming Environments The Dynamic, Exascale Global Address Space programming environment (DEGAS) project will develop the next generation of programming models, runtime systems and tools to meet the challenges of Exascale systems. We will develop a new set of programming concepts based on a hierarchical model of parallelism and data locality, hierarchical fault containment/recovery for resilience, introspective dynamic resource management, demonstrate them using extensions to existing languages, and evaluate…

The FastOS and Tessellation Projects Chip multiprocessors containing hundreds or even thousands of cores will challenge current operating systems (OS) practices. Many of the fundamental assumptions that underlie current OS technology are based on design assumptions that are no longer valid for a chip multiprocessor (CMP) containing thousands of cores. In the context of handheld devices, the OS must manage quality-of-service and resource contention for a complex multi-programmed environment. In the context of high performance…

Intel Parallel Computing Center: Big Data Support on HPC Systems To extend their mission and to open new science frontiers, operators of large scale supercomputers have a vested interest to deploy existing data analytics frameworks such as Spark or Hadoop. So far, this deployment has been hampered by the differences in system architecture, which are reflected in the design approach for the analytics stacks. HPC systems are the mirrored image of data centers. In a data center, the file system is optimized for latency (with local disks) and the network…

Spark Performance on Cray XC with Lustre

PyFloT Tuning floating point precision using dynamic program information and temporal locality.
The goal of this tool is to help finding a mixed precision solution to reduce execution time while maintaining some notion of correctness.