Analysis of SITA policies

Bachmat, Eitan; Sarfati, Hagit

doi:10.1016/j.peva.2009.09.007

“…With FIFO queues, a state-independent dispatching policy known as the size-interval-task-assignment (SITA) has proven to be efficient especially with heavy-tailed job size distributions [6,14,4]. The motivation behind SITA is to segregate the long jobs from the short ones.…”

Section: Sita-e Dispatching Policymentioning

confidence: 99%

Minimizing slowdown in heterogeneous size-aware dispatching systems

Hyytiä

¹

,

Aalto

²

,

Penttinen

³

2012

SIGMETRICS Perform. Eval. Rev.

View full text Add to dashboard Cite

We consider a system of parallel queues where tasks are assigned (dispatched) to one of the available servers upon arrival. The dispatching decision is based on the full state information, i.e., on the sizes of the new and existing jobs. We are interested in minimizing the so-called mean slowdown criterion corresponding to the mean of the sojourn time divided by the processing time. Assuming no new jobs arrive, the shortest-processing-time-product (SPTP) schedule is known to minimize the slowdown of the existing jobs. The main contribution of this paper is three-fold: 1) To show the optimality of SPTP with respect to slowdown in a single server queue under Poisson arrivals; 2) to derive the socalled size-aware value functions for M/G/1-FIFO/LIFO/ SPTP with general holding costs of which the slowdown criterion is a special case; and 3) to utilize the value functions to derive efficient dispatching policies so as to minimize the mean slowdown in a heterogeneous server system. The derived policies offer a significantly better performance than e.g., the size-aware-task-assignment with equal load (SITA-E) and least-work-left (LWL) policies.

show abstract

“…The proofs follow trivially from (7) and from definition (4). Note that ω z (x) with (preemptive) LIFO does not depend on the remaining service times ∆ i due to the preemption.…”

Section: B M/g/1-lifo Queuementioning

confidence: 89%

“…With FIFO queues, a state-independent dispatching policy known as the size-interval-task-assignment (SITA) has proven to be efficient especially with heavy-tailed job size distributions [6], [14], [4]. The motivation behind SITA is to segregate the long jobs from the short ones.…”

Section: B Sita-e Dispatching Policymentioning

confidence: 99%

“…which describes the expected difference in the cumulative sojourn times between System 1 and System 2 for size x jobs. With SRPT also the jobs in System 1 initially longer than x affect the result as eventually, one at a time, they decrease 4 Note that we assume the opposite order than [19]. below the threshold x and trigger a new higher priority workload present only in System 1 (see Fig.…”

Section: B M/g/1-srptmentioning

confidence: 99%

See 1 more Smart Citation

Minimizing slowdown in heterogeneous size-aware dispatching systems

Hyytiä

¹

,

Aalto

²

,

Penttinen

³

2012

Proceedings of the 12th ACM SIGMETRICS/PERFORMANCE Joint International Conference on Measurement and Modeling of Computer Syste

View full text Add to dashboard Cite

We consider a system of parallel queues where tasks are assigned (dispatched) to one of the available servers upon arrival. The dispatching decision is based on the full state information, i.e., on the sizes of the new and existing jobs. We are interested in minimizing the so-called mean slowdown criterion corresponding to the mean of the sojourn time divided by the processing time. Assuming no new jobs arrive, the shortest-processing-time-product (SPTP) schedule is known to minimize the slowdown of the existing jobs. The main contribution of this paper is three-fold: 1) To show the optimality of SPTP with respect to slowdown in a single server queue under Poisson arrivals; 2) to derive the socalled size-aware value functions for M/G/1-FIFO/LIFO/ SPTP with general holding costs of which the slowdown criterion is a special case; and 3) to utilize the value functions to derive efficient dispatching policies so as to minimize the mean slowdown in a heterogeneous server system. The derived policies offer a significantly better performance than e.g., the size-aware-task-assignment with equal load (SITA-E) and least-work-left (LWL) policies.

show abstract

“…In [13] the author introduces the task assignment by guessing size, which is a variant of SITA-E policy where knowledge of the job sizes is not required. Under the SITA routing policy with optimal thresholds, asymptotic analysis for the Bounded Pareto distribution has been done in [4], [26]. The authors in [16] consider a system where the coefficient of variation of incoming tasks is high and they show that the performance of SITA can be much worse than the performance of the Least-Work-Left policy.…”

Section: Related Workmentioning

confidence: 99%

Performance Degradation in Parallel-Server Systems

Doncel

¹

,

Aalto

²

,

Ayesta

³

2019

IEEE/ACM Trans. Networking

View full text Add to dashboard Cite

We consider a parallel-server system with homogeneous servers where incoming tasks, arriving at rate λ, are dispatched by n dispatchers, each of them balancing a fraction 1/n of the load to K/n servers. Servers are FCFS queues and dispatchers implement SITA-E, a size-based policy such that the servers are equally loaded. We compare the performance of a system with n > 1 dispatchers and of a system with a single dispatcher. We show that the performance of a system with n dispatchers, K servers and arrival rate λ coincides with that of a system with one dispatcher, K/n servers and arrival rate λ/n. We define the degradation factor as the ratio between the performance of a system with K servers and arrival rate λ and the performance of a system with K/n servers and arrival rate λ/n. We establish a partial monotonicity on n for the degradation factor and, therefore, the degradation factor is lower-bounded by one. We then investigate the upper-bound of the degradation factor for particular distributions. We consider two continuous service time distributions: uniform and Bounded Pareto; and a discrete distribution with two values, which is the distribution that maximizes the variance for a given mean. We show that the performance degradation is small for uniformly distributed job sizes, but that for Bounded Pareto and two points distributions it can be unbounded. We have investigated the degradation using the distribution obtained from real traces.

show abstract

Analysis of SITA policies

Cited by 28 publications

References 22 publications

Minimizing slowdown in heterogeneous size-aware dispatching systems

Minimizing slowdown in heterogeneous size-aware dispatching systems

Minimizing slowdown in heterogeneous size-aware dispatching systems

Performance Degradation in Parallel-Server Systems

Contact Info

Product

Resources

About