Amazon Technologies, Inc.

Michael Curtis James

Amol Godbole

Prakash Krishnan

Patrick Nevels

Ahmed Reaz

Daniel Rohr

Gavin Hill

Andy Chang

Performance of optimizations to correct duplicate, missing, or other user-induced errors in hierarchical data storage may be automated. Relationships between items may be used to determine whether a modification should be made to items. Once identified, a determination may be made as to whether the modification will impact query performance. If so, then permission may be obtained before the modification is made. If not, then the modification may be made automatically.

Automating optimizations for items in a hierarchical data store

A technology is described for deploying datasets to a production data store. An example of the technology may include receiving a request at a data deployment service to submit a dataset to a production data store which is accessible to services that utilize datasets stored in the production data store. A temporary dataset may be created in a preproduction staging store and a first consistency check may be performed against the temporary dataset to determine whether the temporary dataset complies with dataset specifications. The temporary dataset may be included in a batch of temporary datasets cached on the preproduction staging store when the first consistency check is successful, and a second consistency check may be performed against the batch to determine whether the at least one temporary dataset complies with the dataset specifications. The temporary datasets may be stored to the production data store when the second consistency check is successful.

Validating data for deployment

Growing community of inventors

Secaucus, NJ, United States of America

Michael Curtis James