History-Aware Scheduling for Large Clusters

Student: Korozevtcev Pavel

Supervisor: Ignatiy Kolesnichenko

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2019

Industry uses large clusters to process large amounts of data. Many different operations served by a large number of machines. It allows to share resources between users and to optimize the use of expensive servers. Each process has a quota for the use of resources. We consider the problem of the resource allocation for analytical processes. Such processes do not distribute their use resources evenly over one period. These calculations are run by humans for research purposes. The problem is that a uniform limit resource use does not look fair. As a solution to this problem, we propose a new paradigm of allocation of resources for such processes. In this paper we will consider design of Yandex map-reduce cluster called YT and show how we implement the proposed strategyof restrictions in YT scheduler.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses