Browse SciDAC Centers and Institutes

Development of a collection of algorithmic and software components that can be assembled to simulate a broad range of complex multicomponent physical systems in which partial differential equations plays a central role. Specifically, the goal of APDEC is to enable the agile development of high-performance simulation codes for complex multiphysics and multiscale applications that typically require adaptive mesh refinement.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Louis Howell -
Lawrence Livermore National Laboratory
Ravi Samtaney - Princeton Plasma Physics Laboratory
Ravi Samtaney - Princeton Plasma Physics Laboratory

The Chombo package provides a set of tools for implementing finite difference methods for the solution of partial differential equations on block-structured adaptively refined rectangular grids. Both elliptic and time-dependent modules are included. Support for parallel platforms and standardized self-describing file formats are included.
Capability Areas: [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download
Capability Areas: [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download

Within the context of Chombo, which provides a distributed infrastructure for parallel calculations over block-structured, adaptively refined grids, AMRElliptic is a multigrid-based elliptic equation solver for adaptive hierarchies.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download

Within the context of Chombo, which provides a distributed infrastructure for parallel calculations over block-structured, adaptively refined grids, AMRTimeDependent manages sub-cycling in time for time-dependent adaptive calculations.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Chombo Info
Download

Within the context of Chombo, which provides a distributed infrastructure for parallel calculations over block-structured, adaptively refined grids, AMRTools provides tools for data communication between refinement levels, including coarse-fine interpolation tools.
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download

Within the context of Chombo, which provides a distributed infrastructure for parallel calculations over block-structured, adaptively refined grids, BoxTool provides infrastructure to do any calculations over unions of rectangles and tools to perform set calculus for points in a domain and data holders for unions of rectangles.
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download

Within the context of Chombo, which provides a distributed infrastructure for parallel calculations over block-structured, adaptively refined grids, EBTools provides embedded boundary discretizations and tools.
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Chombo Info
Download

Development of technologies for an efficient and reliable data placement within a distributed high-performance environment and deployment of services for high-performance processing of computation and data analysis requests from many remote clients. The center targets performance and functionality, and focuses on end-to-end concerns, including file system to file system transfers, and integration with user application tools.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Ann Chervenak -
University of Southern California
Dan Gunter - Lawrence Berkeley National Laboratory
Kate Keahey - Argonne National Laboratory
Carl Kesselman - University of Southern California
Mark Livney - University of Wisconsin at Madison
Ruth Pordes - Fermi National Accelerator Laboratory
Dan Gunter - Lawrence Berkeley National Laboratory
Kate Keahey - Argonne National Laboratory
Carl Kesselman - University of Southern California
Mark Livney - University of Wisconsin at Madison
Ruth Pordes - Fermi National Accelerator Laboratory

Services for high-performance, reliable, secure data movement service that is based on the concept of cloud computing. The service can be invoked through a web interface ("drag and drop") or command line-like instructions.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Globus Online Support
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Globus Online Support

Data transfer protocol based on the standard FTP protocol that is optimized for high-bandwidth wide-area networks. By applying a set of changes to FTP, GridFTP enables faster transfer and in-built security, which are essential for the transmission of very large files stemming from applications that use Grid computing resources.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Ian Foster
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Ian Foster
Download

Development of system software, libraries, compilers, and tools to increase the productivity of computational scientists and performance of applications on leadership computing platforms. Topic workshops to engage researchers in the challenges of leadership computing; collaborations with computational scientists for code development on leadership platforms.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Peter Beckman -
Argonne National Laboratory
Jack Dongarra - University of Tennessee at Knoxville
Barton Miller - University of Wisconsin at Madison
Katherine Yelick - Lawrence Berkeley National Laboratory
Jack Dongarra - University of Tennessee at Knoxville
Barton Miller - University of Wisconsin at Madison
Katherine Yelick - Lawrence Berkeley National Laboratory
Co-Array Fortran is a fortran compiler that incorporates a set of features introduced in Fortran 95 that implement a partitioned global address space (PGAS) parallel programming model. These features provide an explicit notation for data decomposition, as an alternative to message-passing models like MPI.
Capability Area: [ Languages and Code Generation ]
Point of Contact: John Mellor-Crummey
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: John Mellor-Crummey
Download


GASNet is the Global-Address Space Networking, a language-independent, low-level networking layer that provides communication primitives tailored for implementing parallel global address space SPMD languages (e.g . UPC).
Capability Area: [ Languages and Code Generation ]
Point of Contact: Berkeley UPC Developers list
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Berkeley UPC Developers list
Download

HPCToolkit is a suite of tools for measurement and analysis of program performance. It uses hardware counters to sample the execution of a multithreaded and/or multiprocess program, enabling fine-grain measurement without modifying the source code
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: HPC
Download
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: HPC
Download

A Java-based tool for graphical analysis of trace files (time-stamped events) and the behavior of parallel programs. It is packed with the MPICH portable implementation of MPI.
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Bill Gropp
Download
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Bill Gropp
Download
LoopTool implements compilation techniques for automatic, source to source loop transformations to improve data reuse and therefore performance. These techniques include fusion of loop nests, array contraction, selective unroll-and-jam and adjustement of loop bounds.
Capability Area: [ Languages and Code Generation ]
Point of Contact: John Mellor-Crummey
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: John Mellor-Crummey
Download
OSKI is the Optimized Sparse Kernel Interface, a collection of low-level C primitives that provide automatically tuned computational kernels on sparse matrices, for use in solver libraries and applications.
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: OSKI Mailing List
Download
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: OSKI Mailing List
Download

PLASMA is the Parallel Linear Algebra for Scalable Multi-core Architectures library uses out-of-order scheduling as the basis for scalable and highly efficient computational linear algebra.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: PLASMA Users Forum
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: PLASMA Users Forum
Download

UPC is Unified Parallel C, a language that provides a uniform programming model for both shared and distributed memory hardware.
Capability Area: [ Languages and Code Generation ]
Point of Contact: UPC Mailing Lists
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: UPC Mailing Lists
Download

ZeptoOS is a research project that focuses on operating systems for petascale architectures. It provides a complete software stack for Blue Gene/P, covering both compute and I/O nodes.
Capability Area: [ Hardware Management ]
Point of Contact: ZeptoOS Users
Download
Capability Area: [ Hardware Management ]
Point of Contact: ZeptoOS Users
Download

The Institute for Combinatorial Scientific Computing and Petascale Simulations (CSCAPES, pronounced "seascapes") seeks to accelerate the development and deployment of fundamental enabling technologies in high performance computing, by creating algorithms and software tools for key combinatorial problems in scientific computing at the petascale. In addition, CSCAPES attempts to foster the next generation of researchers capable of effectively applying combinatorial techniques to scientific computing.
Homepage
Type:
Type:
Institute
Sponsor:
Principal Investigator:
Co-PIs:
Erik Boman -
Sandia National Laboratories
Paul Hovland - Argonne National Laboratory
Ümit Çatalyürek - Ohio State University
Paul Hovland - Argonne National Laboratory
Ümit Çatalyürek - Ohio State University
ADIC is a tool for the automatic differentiation (AD) of programs written in ANSI C. Given the source code and a user's specification of dependent and independent variables, ADIC generates an augmented C code that computes the partial derivatives of all of the specified dependent variables with respect to all of the specified independent variables in addition to the original result.
Capability Area: [ Languages and Code Generation ]
Point of Contact: ADC Questions
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: ADC Questions
Download
ADIFOR is a tool for the automatic differentiation of Fortran 77 programs. Given a Fortran 77 source code and a user's specification of dependent and independent variables, ADIFOR will generate an augmented derivative code that computes the partial derivatives of all of the specified dependent variables with respect to all of the specified independent variables in addition to the original result.
Capability Area: [ Languages and Code Generation ]
Point of Contact: ADIFOR Questions
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: ADIFOR Questions
Download
ColPack is a package of algorithms for the specialized vertex coloring problems as well as algorithms for a variety of related supporting tasks in derivative computation, such as coloring, ordering, recovery and graph construction.
Capability Area: [ Graph Algorithms ]
Point of Contact: Assefaw Gebremedhin
Download
Capability Area: [ Graph Algorithms ]
Point of Contact: Assefaw Gebremedhin
Download
Data_N_Comp_Reorder provides various sparse matrix reordering algorithms that aim to improve the data locality and therefore the performance of computations that iterate over the sparse matrix data structure
Capability Area: [ Graph Algorithms ]
Point of Contact: Paul Hovland
Download
Capability Area: [ Graph Algorithms ]
Point of Contact: Paul Hovland
Download
MatchBox
MatchBox is a C++ software package consisting of sequential exact as well as approximation algorithms for a variety of graph matching problems. The package is currently under development. A trial copy can be obtained by request.
Capability Area: [ Graph Algorithms ]
Point of Contact: Alex Pothen
Capability Area: [ Graph Algorithms ]
Point of Contact: Alex Pothen
MatchBoxP
MatchBoxP provides parallel approximation algorithms for various matching problems implemented in C++ and using MPI. The package is currently under development. A trial copy can be obtained by request.
Capability Area: [ Graph Algorithms ]
Point of Contact: Alex Pothen
Capability Area: [ Graph Algorithms ]
Point of Contact: Alex Pothen

OpenAD is a tool for automatic differentiation (AD) of C/C++ numerical computer programs. For the C/C++ side the entire pipeline is encapsulated in single binary. While the C/C++ tool incarnation shares components with OpenAD its development is a separate effort and it was decided to retain for it the name "ADIC" which goes back to the previous generation of tools.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download

OpenAD/F is a tool for automatic differentiation (AD) of Fortran90 numerical computer programs. The major portion for the Fortran90 front end come from the Open64 compiler in the version that is under development at Rice University.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download
PaToH is the Partitioning Tools for Hypergraph, a multilevel hypergraph partitioning tool that provides fast hypergraph partitioning.
Capability Area: [ Meshing Tools ]
Point of Contact: Ümit V. Çatalyürek
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Ümit V. Çatalyürek
Download
Matlab interface to PaToH, the Partitioning Tools for Hypergraph, a multilevel hypergraph partitioning tool that provides fast hypergraph partitioning.
Capability Area: [ Meshing Tools ]
Point of Contact: Ümit V. Çatalyürek
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Ümit V. Çatalyürek
Download

Rapide Surcharge d'Opérateur pour la Différentiation Automatique, a code generator to efficiently compute higher order derivatives via operator overloading. Rapsodia is intended for the computation of higher order derivative information of numerical models written in Fortran, C or C++ by automatic differentiation.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Jean Utke
Download

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.
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download

Isorropia performs combinatorial scientific computing, with focus on partitioning and load balancing, but also supports coloring and ordering of sparse matrices. Isorropia provides an object-oriented interface to Zoltan using the Epetra linear algebra services.
Capability Area: [ Meshing Tools ]
Point of Contact: Erik Boman
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Erik Boman
Download

Zoltan provides dynamic load balancing for parallel computing with a procedural C interface, with a variety of redistribution algorithms available.
Capability Area: [ Meshing Tools ]
Point of Contact: Vitus Leung
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Vitus Leung
Download

The mission of the DOE SciDAC-2 Earth System Grid Center for Enabling Technologies (ESG-CET) project is to provide climate researchers worldwide with access to data, information, models, analysis tools, and computational resources required to make sense of enormous climate simulation datasets.

The Center for Interoperable Technologies for Advanced Petascale Simulations (ITAPS) delivers interoperable and interchangeable mesh, geometry, and field manipulation services that are of direct use to scientific applications. The premise of our technology development goal is that such services can be provided as libraries that can be used with minimal intrusion into application codes.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Ed D'Azevedo -
Oak Ridge National Laboratory
James Glimm - Brookhaven National Laboratory University of California at Berkeley
Patrick Knupp - Sandia National Laboratories
Xiaolin Li - University of California at Berkeley
Mark Shephard - Rensssealaer Polytechnic Institute
Timothy Tautges - Argonne National Laboratory University of Wisconsin at Madison
Harold Trease - Pacific Northewest National Laboratory
James Glimm - Brookhaven National Laboratory University of California at Berkeley
Patrick Knupp - Sandia National Laboratories
Xiaolin Li - University of California at Berkeley
Mark Shephard - Rensssealaer Polytechnic Institute
Timothy Tautges - Argonne National Laboratory University of Wisconsin at Madison
Harold Trease - Pacific Northewest National Laboratory
Frontier is a C library for fluid-dynamics simulations with front tracking capabilities for surfaces of discontinuity.
Capability Area: [ Meshing Tools ]
Point of Contact: Xiaolin Li
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Xiaolin Li
Download
iGeom is an interface specification for the geometric description of the domains of scientific simulation applications.
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
iGeom interfaces to CGM, the Common Geometry Module, a mesh geometry abstraction layer used by Cubit from Sandia National Laboratories and allows reading a variety of CAD formats.
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download
iGeom interface to MOAB, the Mesh-Oriented Data Base. MOAB can store structured and unstructured mesh information, consisting of elements in the finite element collection, plus polygons and polyhedra.
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
iMesh is an interface specification for the discrete description of the domains of scientific simulation applications.
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download

iMesh interface to FMDB, the Flexible distributed Mesh DataBase. Supports triangles, quadrilaterals, tetrahedron, hexahedron, and prisms, including all upward and downward adjacencies (at varying costs). Supports a fullimplementation of 2D and 3D.
Capability Area: [ Meshing Tools ]
Point of Contact: FMDB-info
Download
Capability Area: [ Meshing Tools ]
Point of Contact: FMDB-info
Download
iMesh interface to Frontier. Supports triangles, quadrilaterals, tetrahedron, hexahedron, all downward adjacenies plus vertex to face upward adjacency, 2D and 3D, partial implementation (set functionality incomplete).
Capability Area: [ Meshing Tools ]
Point of Contact: Xiaolin Li
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Xiaolin Li
Download
iMesh interface to GRUMMP, the Generation and Refinement of Unstructured, Mixed-Element Meshes in Parallel. Supports all mesh topologies except septahedra, polygons, and polyhedra (no edges in 3D), all upward and downward adjacencies (again, no edges in 3D), 2D and 3D, full implementation.
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download
iMesh interface to MOAB, the Mesh-Oriented Data Base. All mesh region topologies supported except septahedron, all upward and downward adjacencies (at varying cost), 2D and 3D, full implementation.
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download
iMesh interface to NWGrid, the NorthWest Grid Generation Code. Supports unstructured mixed elements in 2D and 3D.
Capability Area: [ Meshing Tools ]
Point of Contact: Harold Trease
Capability Area: [ Meshing Tools ]
Point of Contact: Harold Trease
iMesh / RefImpl
iMesh interface to a small, lightweight reference implementation. Derived from GRUMMP, the Generation and Refinement of Unstructured, Mixed-Element Meshes in Parallel, with the same feature set.
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download
iMeshP is a parallel interface specification for the discrete description of the domains of scientific simulation applications.
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download

iMeshP interface to FMDB, the Flexible distributed Mesh DataBase. Supports triangles, quadrilaterals, tetrahedron, hexahedron, and prisms, all upward and downward adjacencies (at varying costs), 2D and 3D, partial implementation. Contact the developers for an alpha version.
Capability Area: [ Meshing Tools ]
Point of Contact: FMDB-info
Capability Area: [ Meshing Tools ]
Point of Contact: FMDB-info
iMeshP interface to MOAB, the Mesh-Oriented Data Base. Supports all mesh region topologies except septahedron, all upward and downward adjacencies (at varying cost), 2D and 3D, full implementation.
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
iRel is an interface specification for managing the relationships between two or more ITAPS interface specifications.
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
Capability Area: [ Meshing Tools ]
Point of Contact: ITAPS Team
Download
iRel / LASSO
iRel interface to LASSO. The Lasso tool implements the iRel interface, working through the other ITAPS interfaces. Lasso can recover the associations between mesh and geometric model entities those mesh elements discretize. Lasso currently works with geometry and mesh read into the CGM and MOAB components
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Timothy Tautges
Download

ITAPS interface for VisIt, a free interactive parallel visualization and graphical analysis tool for viewing scientific data on Unix and PC platforms.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Mark Miller
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Mark Miller
Download
Mesquite is a linkable software library that applies a variety of node-movement algorithms to improve the quality and/or adapt a given mesh. Mesquite uses advanced smoothing and optimization to achieve improved mesh quality in a variety of ways.
Capability Area: [ Meshing Tools ]
Point of Contact: Patrick Knupp
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Patrick Knupp
Download
Swap
Swap performs mesh quality improvement for simplicial meshes via edge and face swapping in 2D and 3D.
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Carl Oliver-Gooch
Download

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.
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download

Isorropia performs combinatorial scientific computing, with focus on partitioning and load balancing, but also supports coloring and ordering of sparse matrices. Isorropia provides an object-oriented interface to Zoltan using the Epetra linear algebra services.
Capability Area: [ Meshing Tools ]
Point of Contact: Erik Boman
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Erik Boman
Download

Zoltan provides dynamic load balancing for parallel computing with a procedural C interface, with a variety of redistribution algorithms available.
Capability Area: [ Meshing Tools ]
Point of Contact: Vitus Leung
Download
Capability Area: [ Meshing Tools ]
Point of Contact: Vitus Leung
Download

Development and promotion of cutting edge visualization technologies to analyze vast amount of complex data produced by scientific simulation applications and to enable knowledge discovery at extreme scales. Engagement with a range of application partners including earthquake simulation, hydrology, climate, astrophysics, fusion, and turbulence.
Homepage
Type:
Type:
Institute
Sponsor:
Principal Investigator:
Co-PIs:
Kenneth Moreland -
Sandia National Laboratories
Ananta Tiwari - University of Maryland at College Park
Greg Humphreys - University of British Columbia
Rob Ross - Argonne National Laboratory
Jian Huang - University of Tennessee at Knoxville
Ananta Tiwari - University of Maryland at College Park
Greg Humphreys - University of British Columbia
Rob Ross - Argonne National Laboratory
Jian Huang - University of Tennessee at Knoxville

CUDPP is the CUDA Data Parallel Primitives Library, a library of data-parallel algorithm primitives such as parallel prefix-sum, parallel sort and parallel reduction.
Capability Area: [ Languages and Code Generation ]
Point of Contact: CUDPP Google Group
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: CUDPP Google Group
Download
DCGN is a message passing interface usable on NVIDIA GPUs that is similar to MPI and allows full access to the underlying architecture, providing both performance and flexibility.
Capability Area: [ Languages and Code Generation ]
Capability Area: [ Languages and Code Generation ]

ParaView is an open-source, multi-platform data analysis and visualization application that allows users to quickly build visualizations to analyze their data using qualitative and quantitative techniques.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: ParaView Users
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: ParaView Users
Download

Development of high performance storage solutions for large scale computer simulations, by bringing together high performance file and storage system expertise. The institute seeks to identify and provide solutions for the storage capacity, performance, concurrency, reliability, availability and manageability for applications at scale.
Homepage
Type:
Type:
Institute
Sponsor:
Principal Investigator:
Co-PIs:
Lee Ward -
Sandia National Laboratories
Phillip Roth - Oak Ridge National Laboratory
John Shalf - Lawrence Berkeley National Laboratory
Gary Grider - Los Alamos National Laboratory
Evan Felix - Oak Ridge National Laboratory
Darrell Long - University of California at Santa Cruz
Han-Wei Shen - Ohio State University
Phillip Roth - Oak Ridge National Laboratory
John Shalf - Lawrence Berkeley National Laboratory
Gary Grider - Los Alamos National Laboratory
Evan Felix - Oak Ridge National Laboratory
Darrell Long - University of California at Santa Cruz
Han-Wei Shen - Ohio State University

PLFS is a parallel virtual file system that rearranges random, concurrent writes into sequential writes to unique files (i.e. N-1 into N-N).
Capability Area: [ Hardware Management ]
Point of Contact: PLFS Development
Download
Capability Area: [ Hardware Management ]
Point of Contact: PLFS Development
Download

Understanding of key factors in scientific codes and computer systems that affect performance; development of models that accurately predict performance; development of tools for performance monitoring, modeling and optimization; development of technology for automatic performance tuning; transfer of performance analysis technology to end users.
Homepage
Type:
Type:
Institute
Sponsor:
Principal Investigator:
Co-PIs:
David Bailey -
Lawrence Berkeley National Laboratory
Jack Dongarra - University of Tennessee at Knoxville
Rob Fowler - University of North Carolina
Mary Hall - University of Utah
Jeff Hollingsworth - University of Maryland at College Park
Peter Honeyman - University of Virginia
John Mellor-Crummey - Rice University
Boyana Norris - Argonne National Laboratory
Allan Snavely - University of California at San Diego
Patrick Worley - Oak Ridge National Laboratory
Bronis de Supinski - Lawrence Livermore National Laboratory
Jack Dongarra - University of Tennessee at Knoxville
Rob Fowler - University of North Carolina
Mary Hall - University of Utah
Jeff Hollingsworth - University of Maryland at College Park
Peter Honeyman - University of Virginia
John Mellor-Crummey - Rice University
Boyana Norris - Argonne National Laboratory
Allan Snavely - University of California at San Diego
Patrick Worley - Oak Ridge National Laboratory
Bronis de Supinski - Lawrence Livermore National Laboratory

Active Harmony provides software architecture that supports distributed execution of computational objects, and allows applications to be tuned with minimal changes to the application and library source code. It can be used to improve the performance of an application during execution taking into account observed performance.
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Ananta Tiwari
Download
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Ananta Tiwari
Download
CHILL is the Composable High-Level Loop Transformation and Parallelization Framework, and provides support for loop transformations (e.g. permutation or loop interchange, tiling or blocking, unrolling or unroll-and-jam, data copy, fusion, distribution, etc).
Capability Area: [ Languages and Code Generation ]
Point of Contact: Chun Chen
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Chun Chen
Download

Orio is an annotation system implemented in Python that enables developers to insert annotations into their source code (in C or Fortran). These annotations are used to trigger performance optimizations on a specified code fragment.
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Boyana Norris
Download
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: Boyana Norris
Download

ROSE is a compiler infrastructure that builds source-to-source program transformation for a variety of programming languages. These transformations can be used for static analysis, program optimization, performance analysis, and cyber-security purposes.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Chunhua "Leo" Liao
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Chunhua "Leo" Liao
Download

Development of advanced data management technologies. The first layer is located immediately on top of hardware, operating systems, file systems, and mass storage systems, and provides parallel data access technology and transparent access to archival storage. The second layer provides indexing, feature identification, and parallel statistical analysis technology. The third layer provides capabilities for the composition of scientific workflows from the components in the previous layer.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Daniel Chavarria-Miranda -
Pacific Northewest National Laboratory
Alok Choudhary - Ohio State University
Chandrika Kamath - Lawrence Livermore National Laboratory
Bertram Ludaescher - University of California at Davis
Rob Ross - Argonne National Laboratory
Nagiza Samatova - Oak Ridge National Laboratory
Claudio Silva - University of Utah
Mladen Vouk - Northwestern University
Alok Choudhary - Ohio State University
Chandrika Kamath - Lawrence Livermore National Laboratory
Bertram Ludaescher - University of California at Davis
Rob Ross - Argonne National Laboratory
Nagiza Samatova - Oak Ridge National Laboratory
Claudio Silva - University of Utah
Mladen Vouk - Northwestern University

FastBit is a library of search functions supported by compressed bitmap indexes. Similarly to other database management systems, Fastbit treats user data in a column-oriented manner but it is designed to accelerate user's data selection tasks.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: FastBit Contacts
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: FastBit Contacts
Download
Kepler is a problem-solving environment for the analysis and modeling of scientific data. It simplifies the creation of executable models by using a visual representation of processes, or “scientific workflows”, which display the flow of data among discrete analysis and modeling components.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Kepler Support
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Kepler Support
Download
pR is lightweight middleware for the R software environment for statistical computing and graphics, and that supports the automatic parallelization of data-parallel tasks in hybrid multi-node and multi-core environments.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Gonzalo Vera Rodriguez
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Gonzalo Vera Rodriguez
Download

PVFS is the Parallel Virtual File System for Linux based clusters that supports a number of access models, from collective I/O to independent I/O and non-contiguous and structured access patterns.
Capability Area: [ Hardware Management ]
Point of Contact: PVFS Support
Download
Capability Area: [ Hardware Management ]
Point of Contact: PVFS Support
Download

Sapphire is a system that implements modular capabilities for scientific data mining, and that targets problems that involve data from observations, experiments and simulations.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Chandrika Kamath
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: Chandrika Kamath

Development of technology to enable the modularization and deployment of large applications, encompassing software components from within and across diverse disciplines, while making a commitment to performance. The center builds upon the Common Component Architecture (CCA) effort, which is a component model targeted to the needs of high-performance scientific computing.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Robert Armstrong -
Sandia National Laboratories
Randall Bramley - Indiana University
Daniel Chavarria-Miranda - Pacific Northewest National Laboratory
Kostadin Damevski - State University of New York at Binghamton
Madhusudhan Govindaraju - North Carolina State University
Lois McInnes - Argonne National Laboratory
Svetlana Shasharina - Tech-X Corporation
Matthew Sottile - University of Oregon
Alan Sussman - University of Maryland at College Park
Randall Bramley - Indiana University
Daniel Chavarria-Miranda - Pacific Northewest National Laboratory
Kostadin Damevski - State University of New York at Binghamton
Madhusudhan Govindaraju - North Carolina State University
Lois McInnes - Argonne National Laboratory
Svetlana Shasharina - Tech-X Corporation
Matthew Sottile - University of Oregon
Alan Sussman - University of Maryland at College Park

The Common Component Architecture is a standard for Component-based software engineering used in high-performance scientific computing. Features of the CCA that distinguish it from commercial component standards include support for FORTRAN, multi-dimensional data arrays, exotic hardware and operating systems, and a variety of network data transports not typically suited for wide area networks.
Capability Area: [ Languages and Code Generation ]
Point of Contact: CCA Forum
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: CCA Forum
Download

Babel is a tool that provides support for language interoperability. It is based on the Scientific Interface Description Language (SIDL), which is specifically tuned for scientific applications and supports complex numbers, structures, and dynamic multidimensional arrays.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Tom Epperly
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Tom Epperly
Download

Ccaffeine is a framework that support component-based HPC application development. it is designed to not interfere with the application after the application has been built from components. It can also automatically generate a static, possibly self-tuning, executable for production runs.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Ben Allan
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Ben Allan
Download

CQoS is the Computational Quality of Service, tools for the automatic selection and configuration of components to suit a particular computational purpose and environment.
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: CQoS Mailing List
Capability Area: [ Performance Analysis and Tuning ]
Point of Contact: CQoS Mailing List

CCA / OnRamp
OnRamp is a mechanism to generate CCA components through commented markup in the original code.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Geoffrey Hulette
Capability Area: [ Languages and Code Generation ]
Point of Contact: Geoffrey Hulette

XCAT is a distributed framework for composing and debbuging applications in the context of CCA, the Common Component Architecture.
Capability Area: [ Languages and Code Generation ]
Point of Contact: Dennis Gannon
Download
Capability Area: [ Languages and Code Generation ]
Point of Contact: Dennis Gannon
Download

The TOPS ("Towards Optimal Petascale Simulations") project responds directly to the needs of scientific applications by developing, demonstrating, and disseminating scalable solver software. The TOPS agenda addresses one of the chief challenges of petascale simulation: solvers whose scaling lags the rest of the simulation, and several other challenges relating to the usability of sophisticated solvers.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
James Demmel -
University of California at Berkeley
Robert Falgout - Lawrence Livermore National Laboratory
Omar Ghattas - University of Texas
Michael Heroux - Sandia National Laboratories
Steve McCormick - University of Colorado at Boulder
Esmond Ng - Lawrence Berkeley National Laboratory
Daniel Reynolds - University of Oregon
Barry Smith - Argonne National Laboratory
Robert Falgout - Lawrence Livermore National Laboratory
Omar Ghattas - University of Texas
Michael Heroux - Sandia National Laboratories
Steve McCormick - University of Colorado at Boulder
Esmond Ng - Lawrence Berkeley National Laboratory
Daniel Reynolds - University of Oregon
Barry Smith - Argonne National Laboratory
Hypre is a library for solving large, sparse linear systems of equations on massively parallel computers using scalable preconditioners.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Robert Falgout
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Robert Falgout
Download
PETSc is a suite of data structures and routines for the parallel, scalable solution of scientific applications modeled by partial differential equations.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: PETSc Maintenance
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: PETSc Maintenance
Download
Prometheus is a multigrid solver for finite element matrices on unstructured meshes in solid mechanics. Part of the PETSc suite of solver tools.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Mark Adams
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Mark Adams
Download
TAO is the Toolkit for Advanced Optimization, focused on the development of software for large-scale optimization problems. Part of the PETSc suite of solver tools.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: TAO Comments
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: TAO Comments
Download

The SUite of Nonlinear and DIfferential/ALgebraic equation Solvers was implemented with the goal of providing robust time integrators and nonlinear solvers that can easily be incorporated into existing simulation codes. The primary design goals were to require minimal information from the user, allow users to easily supply their own data structures underneath the solvers, and allow for easy incorporation of user-supplied linear solvers and preconditioners.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: SUNDIALS Support
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: SUNDIALS Support
Download

CVODE solves initial value problems for ordinary differential equation (ODE) systems. Part of the SUNDIALS SUite of Nonlinear and DIfferential/ALgebraic equation Solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download

CVODES solves ordinary differential equation (ODE) systems and includes sensitivity analysis capabilities (forward and adjoint). Part of the SUNDIALS SUite of Nonlinear and DIfferential/ALgebraic equation Solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download

IDA solves initial value problems for differential-algebraic equation (DAE) systems. Part of the SUNDIALS SUite of Nonlinear and DIfferential/ALgebraic equation Solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download

IDAS solves differential-algebraic equation (DAE) systems and includes sensitivity analysis capabilities (forward and adjoint). Part of the SUNDIALS SUite of Nonlinear and DIfferential/ALgebraic equation Solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download

KINSOL solves nonlinear algebraic systems. Part of the SUNDIALS SUite of Nonlinear and DIfferential/ALgebraic equation Solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Radu Serban
Download
SuperLU is a general purpose library for the direct solution of large, sparse, nonsymmetric systems of linear equations on high performance machines.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: X. Sherry Li
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: X. Sherry Li
Download

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.
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download
Capability Areas: [ Graph Algorithms ] [ Meshing Tools ] [ Solvers and Preconditioners ]
Point of Contact: Trilinos Users Forum
Download

Amesos is the direct sparse solver package in Trilinos, providing clean and consistent interfaces to several third party direct solver libraries, using Epetra for the linear algebra services such as vector and operator definitions.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Ken Stanley
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Ken Stanley
Download

Anasazi is an extensible and interoperable framework for large-scale eigenvalue algorithms. Abstract interface supports Epetra and other linear algebra objects.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Rich Lehoucq
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Rich Lehoucq
Download

AztecOO provides an object-oriented interface the the well-known Aztec Krylov-space iterative solver library. Uses the Epetra linear algebra services to define vector and operator objects.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Michael Heroux
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Michael Heroux
Download

ForTrilinos provides a set of standards-conforming, portable Fortran interfaces to select Trilinos packages.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Damian Rouson
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Damian Rouson
Download

IFPACK provides a suite of object-oriented algebraic preconditioners for the solution of preconditioned iterative solvers.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Michael Heroux
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Michael Heroux
Download

LOCA is the Library of Continuation Algorithms, whose objective is to compute families of solutions to the parameterized nonlinear equation F(x,p) = 0 and their bifurcations.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Roger Pawlowski
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Roger Pawlowski
Download

ML is the Trilinos multigrid preconditioning package designed to solve large sparse linear systems of equations arising primarily from elliptic PDE discretizations.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Jonathan Hu
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Jonathan Hu
Download

NOX is the Nonlinear Object-Oriented Solutions package of Trilinos, which enables the robust and efficient solution of the nonlinear equations of the form F(x) = 0.
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Roger Pawlowski
Download
Capability Area: [ Solvers and Preconditioners ]
Point of Contact: Roger Pawlowski
Download

Leveraging of scientific visualization and analytics software tools as an enabling technology for increasing scientific productivity and insight. The center seeks to respond directly to challenges posed by vast collections of complex data, by adapting, extending, creating when necessary and deploying visualization and data understanding technologies for applications of interest to DOE.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsor:
Principal Investigator:
Co-PIs:
Sean Ahern -
Oak Ridge National Laboratory
Hank Childs - Lawrence Berkeley National Laboratory
Chris Johnson - University of Utah
Ken Joy - University of California at Davis
Ümit Çatalyürek - Ohio State University
Hank Childs - Lawrence Berkeley National Laboratory
Chris Johnson - University of Utah
Ken Joy - University of California at Davis
Ümit Çatalyürek - Ohio State University

ImageVis3D is a volume rendering program developed by the NIH/NCRR Center for Integrative Biomedical Computing with main design goals of simplicity, scalability, and interactivity.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: ImageVis3D Users
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: ImageVis3D Users
Download

SciRun is a problem solving environment for modeling, simulation and visualization of scientific problems; VACET extends the visualization component.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: SCIRun Users
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: SCIRun Users
Download

VisIt is a free interactive parallel visualization and graphical analysis tool for viewing scientific data on Unix and PC platforms
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: VisIt Users
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: VisIt Users
Download

VisTrails is an open-source scientific workflow and provenance management system developed at the University of Utah that provides support for data exploration and visualization.
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: VisTrails Users
Download
Capability Area: [ Visualization and Data Analysis ]
Point of Contact: VisTrails Users
Download

The Open Science Grid (OSG) advances science through open distributed computing. The OSG is a multi-disciplinary partnership to federate local, regional, community and national cyberinfrastructures to meet the needs of research and academic communities at all scales.
Homepage
Type:
Type:
Center for Enabling Technology (CET)
Sponsors:
Principal Investigator: