Name
Abstract | ||
|---|---|---|
An abstract "pebble game" is used to schedule three common families of computation-dags for Internet-based computing (IC, for short): reduction-meshes, which represent (the intertask dependencies of) computations that comprise "up-sweeps" of meshes; reduction-trees, which represent the analogous computations on trees; and FFT dags, which represent a large variety of convolutional computations. The quality of a schedule is assessed by its memory requirements and its rate of producing tasks that are eligible for allocation to clients. These criteria are important because of respectively, the typically enormous sizes of IC computations; the typical temporal unpredictability of clients in IC, which renders a computation vulnerable to stalling indefinitely, pending the return of intermediate results by clients. Under idealized assumptions, the schedules produce eligible tasks at an optimal rate and use memory either exactly or approximately optimally. |
Year | Venue | Field |
|---|---|---|
2005 | PARALLEL AND DISTRIBUTED COMPUTING SYSTEMS | Fixed-priority pre-emptive scheduling,Fair-share scheduling,Computer science,Flow shop scheduling,Two-level scheduling,Rate-monotonic scheduling,Earliest deadline first scheduling,Dynamic priority scheduling,Round-robin scheduling,Distributed computing |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Arnold L. Rosenberg | 1 | 2107 | 640.21 |
| Matthew Yurkewych | 2 | 21 | 1.20 |