StarCluster is a utility for creating traditional computing clusters used in research labs or for general distributed computing applications on Amazon's Elastic Compute Cloud (EC2). It uses a simple configuration file provided by the user to request cloud resources from Amazon and to automatically configure them with a queuing system, an NFS shared /home directory, passwordless SSH, OpenMPI, and ~140GB scratch disk space. It consists of a Python library and a simple command line interface to the library. For end-users, the command line interface provides simple intuitive options for getting started with distributed computing on EC2 (i.e. starting/stopping clusters, managing AMIs, etc). For developers, the library wraps the EC2 API to provide a simplified interface for launching/terminating nodes, executing commands on the nodes, copying files to/from the nodes, etc.
GarlicSim is a platform for writing, running, and analyzing simulations. It is general enough to handle any kind of simulation: physics, game theory, epidemic spread, electronics, etc. GarlicSim aims to eliminate the need to write any boilerplate code that isn't directly related to the phenomenon you're simulating. GarlicSim defines a new format for simulations, called a simulation package and often abbreviated as simpack. The simpack contains all the code that define the simulated system, and is simply a Python package which defines a few special functions according to the GarlicSim simpack API. Simpack code may also be written in C. All of the tools that GarlicSim provides can be used to run simulations of all kinds of different domains.
The Pegasus Workflow Management System encompasses a set of technologies which help workflow-based applications execute in a number of different environments, including desktops, campus clusters, grids, and clouds. It bridges the scientific domain and the execution environment by automatically mapping high-level workflow descriptions onto distributed resources. It automatically locates the necessary input data and computational resources necessary for workflow execution. It enables scientists to construct workflows in abstract terms without worrying about the details of the underlying execution environment or the particulars of the low-level specifications required by the middleware (Condor, Globus, or Amazon EC2). It bridges the current cyberinfrastructure by effectively coordinating multiple distributed resources.