公开(公告)号：US09477523B1

公开(公告)日：2016-10-25

申请号：US13926752

申请日：2013-06-25

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： Leon Robert Warman ,  Mark Austin Buckley ,  Bhavnish H. Lathia ,  Harsha Ramalingam ,  Erik Warren Selberg ,  Robert Eicher Simmering

IPC: G06F9/46 ,  G06F9/48 ,  G06F9/50

CPC classification number: G06F9/4887 ,  G06F9/4843 ,  G06F9/4881 ,  G06F9/5011 ,  G06F9/5038 ,  G06F9/5066 ,  G06F2209/484

Abstract: Techniques are described for scheduling data access jobs based on a job dependency analysis. A requested primary data access job is analyzed to determine one or more preliminary data access jobs on which it depends, and an execution duration of each data access job is predicted based on historical data or other factors. A time-sensitive subset of the preliminary data access jobs is determined as the subset of those serially dependent preliminary data access jobs for which there is a minimum time difference between the total predicted execution duration and a requested target completion time. Data access jobs are scheduled with priority given to those preliminary data access jobs in the time-sensitive subset, to enable the primary data access jobs to be completed by the requested target completion times.

Abstract translation: 描述了基于作业依赖性分析调度数据访问作业的技术。 分析所请求的主数据访问作业以确定其所依赖的一个或多个初步数据访问作业，并且基于历史数据或其他因素来预测每个数据访问作业的执行持续时间。 将初步数据访问作业的时间敏感子集确定为在总预测执行持续时间与所请求的目标完成时间之间存在最小时间差的那些串行依赖的初步数据访问作业的子集。 数据访问作业优先考虑时间敏感子集中的初步数据访问作业，以使主数据访问作业能够通过请求的目标完成时间完成。

公开(公告)号：US10331483B1

公开(公告)日：2019-06-25

申请号：US15286320

申请日：2016-10-05

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： Leon Robert Warman ,  Mark Austin Buckley ,  Bhavnish H. Lathia ,  Harsha Ramalingam ,  Erik Warren Selberg ,  Robert Eicher Simmering

IPC: G06F9/46 ,  G06F9/48 ,  G06F9/50

Abstract: Techniques are described for scheduling data access jobs based on a job dependency analysis. Preliminary data access jobs subsequently accessed by a primary data access job are identified. Execution durations of the primary data access job and the preliminary data access jobs are determined based on historical data. A total execution duration of subsets of serially dependent preliminary data access jobs and predicted execution duration of primary data access job are determined. A time-sensitive subset is identified and start times for the data access job are scheduled. The results of the preliminary data access jobs are provided to the primary data access job prior to the start time of the primary data access job so the primary data access job completes by a target completion duration.