Theoretical Computer Science - Special issue on dynamic and on-line algorithms
Speed is as powerful as clairvoyance
Journal of the ACM (JACM)
Theoretical Computer Science - Selected papers in honor of Manuel Blum
Server scheduling in the Lp norm: a rising tide lifts all boat
Proceedings of the thirty-fifth annual ACM symposium on Theory of computing
Minimizing flow time nonclairvoyantly
Journal of the ACM (JACM)
Non-clair voy ant multiprocessor scheduling of jobs with changing execution characteristics
Journal of Scheduling - Special issue: On-line scheduling
Nonclairvoyant scheduling to minimize the total flow time on single and parallel machines
Journal of the ACM (JACM)
Competitive online scheduling for server systems
ACM SIGMETRICS Performance Evaluation Review
Non-clairvoyant scheduling with precedence constraints
Proceedings of the nineteenth annual ACM-SIAM symposium on Discrete algorithms
Scalably scheduling processes with arbitrary speedup curves
SODA '09 Proceedings of the twentieth Annual ACM-SIAM Symposium on Discrete Algorithms
Non-clairvoyant batch sets scheduling: fairness is fair enough
ESA'07 Proceedings of the 15th annual European conference on Algorithms
Scheduling jobs with varying parallelizability to reduce variance
Proceedings of the twenty-second annual ACM symposium on Parallelism in algorithms and architectures
Brief announcement: online batch scheduling for flow objectives
Proceedings of the twenty-fifth annual ACM symposium on Parallelism in algorithms and architectures
| Hi-index | 0.00 |
We consider the problem of nonclairvoyantly scheduling jobs, which arrive over time and have varying sizes and degrees of parallelizability, with the objective of minimizing the maximum flow. We give essentially tight bounds on the achievable competitiveness. More specifically we show that the competitive ratio of every deterministic nonclairvoyant algorithm is high, namely Ω(√n) for n jobs. But there is a simple batching algorithm that is (1 + ɛ)-processor O(log n)-competitive. And this simple batching algorithm is optimally competitive as no deterministic nonclairvoyant algorithm can be s-processor o(log n)-competitive for any constant s.