Peter Sirota

TL;DR: In this paper, the authors describe techniques for managing distributed execution of programs on a cluster of multiple computing nodes, such as automatically determining the status of execution of the program on each of the multiple nodes and/or automatically determining aggregate usage of one or more types of computing resources across the cluster of many computing nodes by the program execution.

TL;DR: In this article, a definition of a task is given and a need-by-time is defined, which comprises a deadline for completion of execution of the task, and an estimated duration to complete the task is determined for each of a plurality of computing resources.

TL;DR: In this paper, techniques for managing distributed execution of programs, including by dynamically scaling a cluster of multiple computing nodes used to perform ongoing distributed execution, such as to increase and/or decrease the quantity of computing nodes in the cluster at various times and for various reasons.

TL;DR: In this article, the authors describe techniques for managing distributed execution of programs, such as dynamically modifying the distributed program execution in various manners, based on monitored status information, which may include adding and/or removing computing nodes from a cluster that is executing the program.