US 12,475,081 B2
Adaptive archive of data chunkfiles
Deepak Ojha, Fremont, CA (US); and Zhihuan Qiu, Santa Clara, CA (US)
Assigned to Cohesity, Inc., San Jose, CA (US)
Filed by Cohesity, Inc., San Jose, CA (US)
Filed on Jul. 25, 2023, as Appl. No. 18/358,585.
Prior Publication US 2025/0036593 A1, Jan. 30, 2025
Int. Cl. G06F 16/11 (2019.01); G06F 11/34 (2006.01); G06F 16/174 (2019.01)
CPC G06F 16/113 (2019.01) [G06F 11/3419 (2013.01); G06F 16/1752 (2019.01)] 15 Claims
OG exemplary drawing
 
1. A method of adaptive data archive comprising:
receiving from an application system, by a data platform implemented by a computing system, a stream of native format data;
predicting, by the data platform, an incoming data rate of the stream of native format data to be archived by the data platform to produce a predicted incoming data rate;
responsive to determining the predicted incoming data rate satisfies a first threshold data rate:
segmenting the stream of native format data into one or more chunks;
causing an archive storage system to store the one or more chunks within one or more chunkfiles; and
responsive to determining the predicted incoming data rate satisfies a second threshold data rate after determining the predicted incoming data rate fails to satisfy the first threshold data rate and a time limit for writing the native format data to a local repository has been exceeded:
replaying receipt of the stream of the native format data;
segmenting the stream of native format data into one or more chunks;
writing the one or more chunks to a local repository of the data platform, wherein writing the native format data to the local repository comprises writing the native format data to one of a random access memory or a flash memory of the local repository;
generating the one or more chunkfiles with that includes a copy of the one or more chunks; and
writing the one or more chunkfiles to cause the archive storage system to store the one or more chunkfiles.