Publications

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Recovery from process loss during the execution of a distributed memory parallel application is presently achieved by restarting the program, typically from a checkpoint file. Future computer system trends indicate that the size of data to checkpoint, the lack of improvement in parallel file system performance and the increase in process failure rates will lead to situations where checkpoint restart becomes infeasible. In this report we describe and prototype the use of a new application level resilient computing model that manages persistent storage of local state for each process such that, if a process fails, recovery can be performed locally without requiring access to a global checkpoint file. LFLR provides application developers with an ability to recover locally and continue application execution when a process is lost. This report discusses what features are required from the hardware, OS and runtime layers, and what approaches application developers might use in the design of future codes, including a demonstration of LFLR-enabled MiniFE code from the Matenvo mini-application suite.

Abstract not provided.

Krylov subspace projection methods are widely used iterative methods for solving large-scale linear systems of equations. Researchers have demonstrated that communication avoiding (CA) techniques can improve Krylov methods' performance on modern computers, where communication is becoming increasingly expensive compared to arithmetic operations. In this paper, we extend these studies by two major contributions. First, we present our implementation of a CA variant of the Generalized Minimum Residual (GMRES) method, called CAGMRES, for solving no symmetric linear systems of equations on a hybrid CPU/GPU cluster. Our performance results on up to 120 GPUs show that CA-GMRES gives a speedup of up to 2.5x in total solution time over standard GMRES on a hybrid cluster with twelve Intel Xeon CPUs and three Nvidia Fermi GPUs on each node. We then outline a domain decomposition framework to introduce a family of preconditioners that are suitable for CA Krylov methods. Our preconditioners do not incur any additional communication and allow the easy reuse of existing algorithms and software for the sub domain solves. Experimental results on the hybrid CPU/GPU cluster demonstrate that CA-GMRES with preconditioning achieve a speedup of up to 7.4x over CAGMRES without preconditioning, and speedup of up to 1.7x over GMRES with preconditioning in total solution time. These results confirm the potential of our framework to develop a practical and effective preconditioned CA Krylov method.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The High Performance Conjugate Gradient (HPCG) benchmark [cite SNL, UTK reports] is a tool for ranking computer systems based on a simple additive Schwarz, symmetric Gauss-Seidel preconditioned conjugate gradient solver. HPCG is similar to the High Performance Linpack (HPL), or Top 500, benchmark [1] in its purpose, but HPCG is intended to better represent how today’s applications perform. In this paper we describe the technical details of HPCG: how it is designed and implemented, what code transformations are permitted and how to interpret and report results.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The High Performance Linpack (HPL), or Top 500, benchmark [1] is the most widely recognized and discussed metric for ranking high performance computing systems. However, HPL is increasingly unreliable as a true measure of system performance for a growing collection of important science and engineering applications. In this paper we describe a new high performance conjugate gradient (HPCG) benchmark. HPCG is composed of computations and data access patterns more commonly found in applications. Using HPCG we strive for a better correlation to real scientific application performance and expect to drive computer system design and implementation in directions that will better impact performance improvement.

The Trilinos Project is an effort to develop algorithms and enabling technologies within an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. A new software capability is introduced into Trilinos as a package. A Trilinos package is an integral unit and, although there are exceptions such as utility packages, each package is typically developed by a small team of experts in a particular algorithms area such as algebraic preconditioners, nonlinear solvers, etc. The Trilinos Developers SQE Guide is a resource for Trilinos package developers who are working under Advanced Simulation and Computing (ASC) and are therefore subject to the ASC Software Quality Engineering Practices as described in the Sandia National Laboratories Advanced Simulation and Computing (ASC) Software Quality Plan: ASC Software Quality Engineering Practices Version 3.0 document [1]. The Trilinos Developer Policies webpage [2] contains a lot of detailed information that is essential for all Trilinos developers. The Trilinos Software Lifecycle Model [3] defines the default lifecycle model for Trilinos packages and provides a context for many of the practices listed in this document.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The push to exascale computing is informed by the assumption that the architecture, regardless of the specific design, will be fundamentally different from petascale computers. The Mantevo project has been established to produce a set of proxies, or 'miniapps,' which enable rapid exploration of key performance issues that impact a broad set of scientific applications programs of interest to ASC and the broader HPC community. Understanding the conditions under which a miniapp can be confidently used as predictive of an applications' behavior must be clearly elucidated. Toward this end, we have developed a methodology for assessing the predictive capabilities of application proxies. Adhering to the spirit of experimental validation, our approach provides a framework for examining data from the application with that provided by their proxies. In this poster we present this methodology, and apply it to three miniapps developed by the Mantevo project. © 2012 IEEE.

Abstract not provided.

With the ubiquity of multicore processors, it is crucial that solvers adapt to the hierarchical structure of modern architectures. We present ShyLU, a "hybrid-hybrid" solver for general sparse linear systems that is hybrid in two ways: First, it combines direct and iterative methods. The iterative part is based on approximate Schur complements where we compute the approximate Schur complement using a value-based dropping strategy or structure-based probing strategy. Second, the solver uses two levels of parallelism via hybrid programming (MPI+threads). ShyLU is useful both in shared-memory environments and on large parallel computers with distributed memory. In the latter case, it should be used as a sub domain solver. We argue that with the increasing complexity of compute nodes, it is important to exploit multiple levels of parallelism even within a single compute node. We show the robustness of ShyLU against other algebraic preconditioners. ShyLU scales well up to 384 cores for a given problem size. We also study the MPI-only performance of ShyLU against a hybrid implementation and conclude that on present multicore nodes MPI-only implementation is better. However, for future multicore machines (96 or more cores) hybrid/ hierarchical algorithms and implementations are important for sustained performance. © 2012 IEEE.

Abstract not provided.