Abstractions for cloud computing with condor

A cloud computer provides a simple interface that allows end users to allocate large amounts of computing power and storage space at the touch of a button. However, many potential users of cloud computers have needs much more complex than simply the ability to allocate resources. In scientific domains, it is easy to find examples of workloads that consist of hundreds or thousands of interacting processes. A user who wishes to run such a workload on a cloud computer faces the daunting task of deciding how many resources to allocate, where to dispatch each process, when and where to move data, and how to deal with the inevitable failures. For this reason, many users with large workloads are reluctant to move away from the predictable environment of a single workstation or multicore server.