On-line load balancing

Azar, Yossi; Bröder, Arndt; Karlin, Anna R.

doi:10.1016/0304-3975(94)90153-8

Cited by 83 publications

(62 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The natural greedy algorithm that always chooses the least loaded server is (2 − 1/m)-competitive in the identical machines model [14] and (log m)-competitive in the restricted assignment model [4]. In the temporal tasks model [2], [3], tasks have a limited duration (which is unknown to the on-line algorithm until departure) and disappear after they are finished. In this model the greedy algorithm is only (m 2/3 )-competitive [2].…”

mentioning

confidence: 99%

Balanced Scheduling toward Loss-Free Packet Queuing and Delay Fairness

Fleischer

Koga

2003

Algorithmica

View full text Add to dashboard Cite

In current networks, packet losses can occur if routers do not provide sufficiently large buffers. This paper studies how many buffers should be provided in a router to eliminate packet losses. We assume a network router has m incoming queues, each corresponding to a single traffic stream, and must schedule at any time on-line from which queue to take the next packet to send out. To exclude packet losses with a small amount of buffers, the maximum queue length must be kept low over the entire scheduling period. We call this new on-line problem the balanced scheduling problem (BSP). By competitive analysis, we measure the power of on-line scheduling algorithms to prevent packet losses. We show that a simple greedy algorithm is (log m)-competitive which is asymptotically optimal, while Round-Robin scheduling is not better than m-competitive, as actually is any deterministic on-line algorithm for BSP. We also give a polynomial time algorithm for solving off-line BSP optimally.We also study another on-line balancing problem that tries to balance the delay among the m traffic streams.

show abstract

mentioning

confidence: 99%

Balanced Scheduling toward Loss-Free Packet Queuing and Delay Fairness

Fleischer

Koga

2003

Algorithmica

View full text Add to dashboard Cite

show abstract

“…(Again, under this definition the ratio is 2 − ε if the eligible subset is all servers.) In both [3] and [4] the same bounds are shown for an extension where tasks have weights. These strong bounds indicate that the performance of an on-line scheduler may be severely impacted by the lack of information about future tasks.…”

Section: Load Balancing With Preemptionmentioning

confidence: 52%

“…Azar et al [3] proposed the dynamic load balancing problem as an abstraction of scheduling problems that can occur in heterogenous networks containing workstations, I/O devices, etc. Servers correspond to communication channels and tasks to requests for communication links between devices.…”

Section: Load Balancing With Preemptionmentioning

confidence: 99%

“…Azar et al [3] studied the dynamic load balancing problem when tasks both arrive and depart, but retained the restriction that tasks cannot be preempted. In this version the picture is worse: using our definition of Load(A, σ ), the best possible competitive ratio is n. Azar et al defined Load(A, σ ) to be the maximum load at any time during σ and showed under this definition that the greedy strategy has a ratio of (n 2/3 ).…”

Section: Load Balancing With Preemptionmentioning

confidence: 99%

“…First, we may never preempt and use the algorithm of Azar et al [3]. This gives a ratio of (n 2/3 ) and optimal computation time linear in the total degree.…”

Section: Load Balancing With Preemptionmentioning

confidence: 99%

See 2 more Smart Citations

On-Line Load Balancing and Network Flow

Phillips

Westbrook

1998

Algorithmica

View full text Add to dashboard Cite

In this paper we study two problems that can be viewed as on-line games on a dynamic bipartite graph. The first problem is on-line load balancing with preemption. A centralized scheduler must assign tasks to servers, processing on-line a sequence of task arrivals and departures. Each task is restricted to run on some subset of the servers. The scheduler attempts to keep the load well-balanced. If preemptive reassignments are disallowed, Azar et al.[3] proved a lower bound of ( √ n) on the ratio between the maximum load achieved by an on-line algorithm and the optimum off-line maximum load. We show that this ratio can be greatly reduced by an efficient scheduler using only a small amount of rescheduling.We then apply these ideas to network flow. Cheriyan and Hagerup [6] introduced an on-line game on a bipartite graph as a fundamental step in improving algorithms for computing the maximum flow in networks. They described a randomized strategy to play the game. King et al.[11] studied a modified version of this game, called "node kill," and gave a deterministic strategy. We obtain an improved deterministic algorithm for the node kill game (and hence for maximum flow) in all but the sparsest graphs. The running time achieved is O(mn log m/n n + n 2 log 2+ε n), compared with King et al.'s O(mn + n 2+ε ).These problems combine a demand for good competitive ratios with more traditional requirements of implementation efficiency. Our solutions deal with the tradeoffs between these measures.

show abstract

New algorithms for related machines with temporary jobs

2000

View full text Add to dashboard Cite

We consider the on‐line problem of assigning temporary jobs to related machines. In this model, machines have speeds and the jobs are weighted and may be temporary (i.e. may expire after some unknown finite amount of time). The cost of an assignment is the maximum load on a machine at any time. We present two new algorithms: a 6+2√5≈10.47‐competitive deterministic algorithm and a 9.572‐competitive randomized algorithm. The previously known best upper bound is achieved by a 20‐competitive deterministic algorithm whose randomized version is 5e=13.59‐competitive. Copyright © 2000 John Wiley & Sons, Ltd.

show abstract

On-line load balancing

Cited by 83 publications

References 7 publications

Balanced Scheduling toward Loss-Free Packet Queuing and Delay Fairness

Balanced Scheduling toward Loss-Free Packet Queuing and Delay Fairness

On-Line Load Balancing and Network Flow

New algorithms for related machines with temporary jobs

Contact Info

Product

Resources

About