64 research outputs found
Aceetaldehyde
No full textDocument(en) uit de collectie Chemische Procestechnologie.DelftChemTechApplied Science
Analysis of large scale randomized load balancing policies
No full textAbstract: In the load balancing literature, there has been a lot of interest in policies which balance load based on some information of randomly selected servers. The analysis of these policies for a finite system is intractable in most cases, therefore one often relies on mean field methods. That is, you assume that all queues are i.i.d.~which allows to describe the system's behaviour through the behaviour of a single queue. Most prior work has focused on load balancing methods which balance the load based on the queue length. While reducing response times, small jobs may still get stuck behind long jobs for these policies. In contrast, our main focus is on policies which employ the actual work present at the servers. While causing additional overhead, these policies do allow to detect all large jobs. This includes policies which either favour servers with a low load, use some form of redundancy, have a memory at the dispatcher, assign jobs in batches,... We additionally consider policies which combine the queue length information with the runtime of the job currently receiving service. This allows us to detect large jobs if they have been receiving service for some time. For all policies considered we obtain a numerical method to study its performance. We make use of simulations to illustrate the accuracy of the mean field approximation. Moreover, assuming exponential job sizes often allows us to compute performance metrics in closed form. In particular we noticed that by using a proper scaling, we can often obtain simple closed form formulas for the expected waiting time in case the system load is either close to instability or close to zero
On the power-of-d-choices with least loaded server selection
No full textAbstract: Motivated by distributed schedulers that combine the power-of-d-choices with late binding and systems that use replication with cancellation-on-start, we study the performance of the LL(d) policy which assigns a job to a server that currently has the least workload among d randomly selected servers in large-scale homogeneous clusters. We consider general job size distributions and propose a partial integro-differential equation to describe the evolution of the system. This equation relies on the earlier proven ansatz for LL(d) which asserts that the workload distribution of any finite set of queues becomes independent of one another as the number of servers tends to infinity. Based on this equation we propose a fixed point iteration for the limiting workload distribution and study its convergence. For exponential job sizes we present a simple closed form expression for the limiting workload distribution that is valid for any work-conserving service discipline as well as for the limiting response time distribution in case of first-come-first-served scheduling. We further show that for phase-type distributed job sizes the limiting workload and response time distribution can be expressed via the unique solution of a simple set of ordinary differential equations. Numerical and analytical results that compare response time of the classic power-of-d-choices algorithm and the LL(d) policy are also presented and the accuracy of the limiting response time distribution for finite systems is illustrated using simulations
Warmte-overdracht in een roterende xanthogeneerreactor: Metingen aan twee xanthogeneerreactoren
No full textApplied SciencesKramers Laboratorium voor Fysische Technologi
Phenotype data (composition, functionality, end-product quality) of wheat grown in Flanders, Belgium
No full textData for the Ph.D. dissertation entitled: Outline of four degrees of diversification for understanding bread wheat (Triticum aestivum L.) quality. For further details on the content of the files, the meaning of abbreviations and used techniques, it is referred to the Ph.D. manuscript.
For further details, please contact the author
Improved load balancing in large scale systems using attained service time reporting
No full textAbstract: Our interest lies in load balancing jobs in large scale systems consisting of multiple dispatchers and FCFS servers. In the absence of any information on job sizes, a popular load balancing method is the SQ(d) policy, which uses queue length information reported by the servers to assign incoming jobs. When job sizes are highly variable, using only queue length information is clearly suboptimal and performance can be improved if some indication can be provided to the dispatcher about the size of an ongoing job. In a FCFS server measuring the attained service time of the ongoing job is easy and servers can therefore report this attained service time together with the queue length when queried by a dispatcher. In this paper, we propose and analyse a variety of load balancing policies which improve the SQ(d) policy by exploiting both the queue length and attained service time to assign jobs, as well as policies for which only the attained service time of the job in service is used. We present a unified analysis for all these policies in a large scale system under the usual asymptotic independence assumptions. The accuracy of the proposed analysis is illustrated using simulation. We present extensive numerical experiments which clearly indicate that a significant improvement in waiting (and thus also in response) time may be achieved by using the attained service time information on top of the queue length of a server. Moreover, the policies which do not make use of the queue length still provide an improved waiting time for moderately loaded systems
11. An Arrangement of Silence
Get PDFThe importance of silence for contemplation in the medieval monastery is well known. This chapter examines an alternative silence - that of withholding information. Like many letter collections of the twelfth century, those ascribed to Anselm of Canterbury (1033-1109) may have presented a selective image of their author. The lacuna in the documentation of Anselm’s first political exile raises the question of the extent to which monastic silence and the act of letter collecting are intertwined. Thus I analyse the arrangements of two such letter collections from the perspective of a ‘macro-text’ - an independent narrative. More specifically, I argue that these collections render an image of Anselm that idealizes silence as the virtue of an archbishop living in a state of monastic renunciation.</p
- …
