The current NSF Middleware Initiative (NMI) release is NMI-R5, which is available now for
download. NMI-R5 includes the open-source GRIDS Center
Software Suite, which is aimed at the national research, education,
and scientific communities. See this separate
technology
overview for details about the following GRIDS components.
The Globus Toolkit® from the Globus Alliance, led
by Argonne National Laboratory, the University of Southern California's
Information Sciences Institute and the University of Chicago. This set of
software tools and services is the
foundation on which users may build Grids and Grid-based applications. It has
three key functions: Resource Management to allocate shared computers, storage,
sensors, instruments, networks and software; Information Services to
characterize these shared resources; and Data Management to let users access and
organize information generated by resources.
Condor-G is from the University of Wisconsin at
Madison. It manages both a queue of jobs and a set of resources from one or more
sites where those jobs can execute. Condor-G lets the user submit many jobs at
once and monitor them with a convenient interface. It notifies the users when
jobs complete or fail, and it maintains Globus Toolkit credentials that may
expire while a job is running. Condor-G is also fault-tolerant -- if a machine
crashes, the user can still perform all of these functions after the system
reboots.
The Network Weather Service (NWS), from the
University of California Santa Barbara, is a distributed, generalized system for
producing short-term performance forecasts based on historical performance
measurements. It features a Name Server directory to bind process and data names
with low-level contact information; a Memory Server for persistent data storage;
a Sensor to gather performance measurements from a resource; and a Forecaster to
predict performance of a resource over specified times.
GSI-OpenSSH is a modified version of OpenSSH that
adds support for Grid Security Infrastructure (GSI) authentication. Developed by
the National Center for Supercomputing Applications (NCSA), it can be used to
log into remote systems and transfer files between systems without entering a
password. Instead, all operations can be authenticated with a valid GSI
credential. GSI-OpenSSH can forward GSI credentials to a remote system on login,
so the user is not required to manually create a new credential on that system.
MyProxy is a credential repository from NCSA that
lets Grid users retrieve a proxy credential on demand, without having to manage
private key and certificate files. MyProxy can improve security and flexibility,
so job submissions will not fail due to expired credentials.
MPICH-G2, from Northern Illinois University and
Argonne, is a Grid-enabled imple-mentation of the Message Passing Index (MPI)
standard that is based on the popular MPICH library. It works with the Globus
Toolkit to link multiple machines running MPI appli-cations, even with different
architectures.
NCSA's Grid Packaging Tools (GPT), provide a
straightforward way to define complex dependency and compatibility relationships
between packages. Installation of the software is simplified due to bundling
with GPT, which lets users choose whether to install components collectively or
individually.
The GRIDS Center suite includes KX.509, a client
tool developed at the University of Michigan under the auspices of a partner NMI
team, EDIT (Enterprise and Desktop Integration Technologies). It lets sites
easily convert Kerberos certificates to the X.509 format used by the Grid.
Netscape and Internet Explorer will accept these credentials for secure https
web activity.
GridConfig from the San Diego Supercomputer
Center (SDSC) is used to configure and finetune Grid technologies. It provides
an easy way to generate and regenerate configuration files in native formats,
and to ensure consistency across applications.
GridSolve, from the University of Tennessee, uses
the remote procedure call (RPC) protocol to create a client/agent/server system
for remote access to Grid-enabled hardware and software.
PyGlobus, from Lawrence Berkeley National
Laboratory, permits users to access the Globus Toolkit from Python, a high-level
scripting language.
UberFTP, from the National Center for Supercomputing
Applications (NCSA), is an interactive client for GridFTP, which is part of the
Globus Toolkit
GridPort, from TACC, enables development of portals
and applications on top of underlying distributed and grid computing
infrastructure to facilitate computational science. GridPort provides a
comprehensive set of capabilities for using distributed computing resources via
a consistent API that presents streamlined access to backend grid services from
diverse grid computing technologies. (New in NMI-R5)
DataCutter, from Ohio State University and the
University of Maryland, is a framework designed to provide support for
processing of large scientific datasets in heterogeneous environments. It
supports a filter-stream programming model for executing application-specific
data processing, enabling combined use of task- and data-parallelism. (New in
NMI-R5)
DataCutter STORM, also from OSU and UMD, is a
services-based middleware is designed to support data select and transfer
operations on large and distributed scientific datasets. The objective of STORM
is to enable execution of select queries on datasets stored in files distributed
across a network. (New in NMI-R5)
AppLeS Parameter Sweep Template (APST), from SDSC,
is a tool that schedules and deploys parameter sweep applications on the
Computational Grid. Its purpose is to schedule and deploy parameter sweep
applications on the Computational Grid. Common examples include all kinds of
Monte-Carlo simulations and parameter-space searches. (New in NMI-R5)
INCA, from TeraGrid, is a generic framework
automates testing, verification, and monitoring of functionality common to a set
of Grid systems. (New in NMI-R5)
Storage Resource Broker (SRB), from SDSC, is
client-server middleware that provides a uniform interface for connecting to
heterogeneous data resources over a network and accessing replicated data sets.
SRB, in conjunction with the Metadata Catalog (MCAT), provides a way to access
data sets and resources based on their attributes and/or logical names rather
than their names or physical locations. (New in NMI-R5)
Acknowledgements
Primary funding for GRIDS is from the National Science Foundation
(NSF) Middleware Initiative program 03-513. GRIDS software
developers wish also to acknowledge support from:
NSF Partnerships for
Advanced Computational Infrastructure (for Globus Toolkit,
Condor-G, NWS, MyProxy, GSI-OpenSSH, GPT and GridConfig)
U.S. Department of
Energy (for Globus Toolkit, Condor-G, NWS and MPICH-G2)
Defense Advanced
Research Projects Agency (for Globus Toolkit and Condor-G)
NASA (for Globus
Toolkit, MyProxy and GSI-OpenSSH)
IBM (for Globus
Toolkit)
Microsoft Corporation
(for Globus Toolkit and Condor-G)