公开(公告)号：US12204521B2

公开(公告)日：2025-01-21

申请号：US16905813

申请日：2020-06-18

Applicant: NETFLIX, INC.

Inventor： Daniel Watson ,  Arun Agrawal ,  Charles Smith

IPC: G06F16/23 ,  G06F11/14 ,  G06F16/174 ,  G06F16/178

Abstract: In various embodiments, a consistency application constructs a consistent view of an eventually consistent database. The consistency application determines multiple backup files that are associated with at least one datacenter included in the eventually consistent database and extracts aggregated data from the backup files. The consistency application performs compaction operation(s) on the aggregated data to generate compacted data. Notably, the aggregated data includes at least two replicas for each data item stored in the eventually consistent database, whereas the compacted data includes a different consistent data item for each data item stored in that eventually consistent database. The consistency application generated the consistent view of the eventually consistent database based on the compacted data. Because the consistency application generates the consistent view based on backup files and does not access the eventually consistent database, generating the consistent view does not adversely impact the performance of the eventually consistent database.

公开(公告)号：US10942812B2

公开(公告)日：2021-03-09

申请号：US15476926

申请日：2017-03-31

Applicant: NETFLIX, INC.

Inventor： Charles Smith ,  Jeffrey Magnusson ,  Siddharth Anand

IPC: G06F12/00 ,  G06F11/14 ,  G06F16/23 ,  G06F16/2458 ,  G06F11/16 ,  G06F11/20

Abstract: A method and system for building a point-in-time snapshot of an eventually-consistent data store. The data store includes key-value pairs stored on a plurality of storage nodes. In one embodiment, the data store is implemented as an Apache® Cassandra database running in the “cloud.” The data store includes a journaling mechanism that stores journals (i.e., inconsistent snapshots) of the data store on each node at various intervals. In Cassandra, these snapshots are sorted string tables that may be copied to a back-up storage location. A cluster of processing nodes may retrieve and resolve the inconsistent snapshots to generate a point-in-time snapshot of the data store corresponding to a lagging consistency point. In addition, the point-in-time snapshot may be updated as any new inconsistent snapshots are generated by the data store such that the lagging consistency point associated with the updated point-in-time snapshot is more recent.