Dynamic File Allocation on a Non-Shared Disk Cluster

In data intensive applications, the bottleneck is in getting the data to the processing system. Typically, the amount of processing per unit of data is small. In HENP applications, the analysis of the data can be trivially processed in parallel, since each event can be processed independently of other events.

Dynamic File Allocation on a Non-Shared Disk Cluster

Blogged with Flock