System and method for cluster management

Patent metadata

Field	Value
Publication number	US-9104493-B2
Application number	US-201213670199-A
Country	US
Kind code	B2
Filing date	Nov 6, 2012
Priority date	Nov 6, 2012
Publication date	Aug 11, 2015
Grant date	Aug 11, 2015

Abstract

A system and method of managing a cluster of distributed machines is described. A cluster manager receives status updates regarding tasks running on each machine in the cluster from a task tracker running on the machine. The cluster manager receives resource requests from a job tracker created by a client wishing to run a job in the cluster. The cluster manager is responsible for implementing push-based fair scheduling of resources to the job trackers. The job tracker is responsible for running tasks for one job in the resource identified by the cluster manager. In one embodiment, the job tracker can run in the client for small jobs and in the cluster for larger jobs. The cluster manager can also be restarted, for example, for software updates without restraining the cluster.

First claim

We claim:

1. A system, comprising:
a machine cluster comprising a plurality of server computing devices for running tasks;
a cluster manager configured to track available resources in the machine cluster; and
a job tracker configured to submit to the cluster manager a resource request for a task to be run in the machine cluster, wherein the job tracker is created at a client computing device external to the machine cluster and runs at the client computing device;
wherein the…

Assignees

• Facebook Inc

Inventors

• Molkov Dmytro
• Vadali Ramkumar Venkat
• Chen Chun-Yang
• Sarma Joydeep Sen

Classifications

• G06F2209/5013

  Physics · mapped topic

• G06F9/5072 Primary

  Physics · mapped topic