主权项 |
1. In a network computing environment, a method of moving data from a first system to a second system remote from the first system and performing computations on the data, the method comprising:
at a scheduler of the first system, receiving from one or more users a plurality jobs, each of the one or more jobs specifying both (i) data to be operated on, and (ii) one or more operations to be performed on the data at the second system, the second system being remote from the first system; at a data service of the first system, selecting one of the jobs in the plurality of jobs to upload to the second system in priority over other jobs in the plurality of jobs, and beginning to upload the prioritized job to the second system, including uploading a portion of the data for the prioritized job for staging at the second system; at the scheduler, receiving notice that the portion of data for the prioritized job has been staged at the second system; and based on receiving the notice, at the scheduler, instructing the second system to begin performing one or more operations for the prioritized job on the portion of the data for the prioritized job, where after the second system begins to perform the one or more operations for the prioritized job on the portion of data for the prioritized job, wherein beginning to perform the one or more operations is performed prior to all of the data for the prioritized job being uploaded to the second system. |