US 12,423,363 B2
Warm tier storage for search service
Ramakrishna Kotla, Cupertino, CA (US); Eli R Fisher, San Francisco, CA (US); Min Zhou, Sunnyvale, CA (US); Ankit Malpani, Santa Clara, CA (US); Ankit Kumar Jain, Mountain View, CA (US); Andi Gutmans, Menlo Park, CA (US); and Bhaven Avalani, Cupertino, CA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Apr. 27, 2020, as Appl. No. 16/859,923.
Claims priority of provisional application 62/942,037, filed on Nov. 29, 2019.
Prior Publication US 2021/0165840 A1, Jun. 3, 2021
Int. Cl. G06F 16/953 (2019.01); G06F 16/951 (2019.01); G06F 9/455 (2018.01)
CPC G06F 16/953 (2019.01) [G06F 16/951 (2019.01); G06F 9/45533 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implement method, comprising:
receiving, by a search service, customer data from a customer of a computing resource service provider;
indexing, by the search service, the customer data to produce an index by a first pool of hot compute nodes;
storing, by the search service, the customer data and the index in the first pool of hot compute nodes, individual hot compute nodes of the first pool of hot compute nodes being responsible locally for portions of the customer data and associated portions of the index,
maintaining, by the search service, a first portion of the customer data of the portions of the data in the first pool of hot compute nodes until the migration conditions are satisfied, the migration conditions comprising the first portion of data being stored in the first pool of hot compute nodes for more than a threshold amount of time and fewer than a threshold number of requests to access the first portion of the customer data having been received; and
after the migration conditions are satisfied:
transmitting, by the search service, the first portion of the customer data and the index to a data storage service to be stored remotely thereby; and
causing, by the search service, metadata associated with at least one of the first portion of the customer data and the index to be stored in a second pool of warm compute nodes provisioned by the search service for the customer; and
upon receiving a request to access the first portion of the customer data:
identifying a location of the first portion of customer data;
automatically filtering the first portion of customer data to remove data of the first portion of customer data that corresponding to timestamp information that is beyond a threhsold age, the time stamp information contained in the first portion of customer data; and
retrieving the first portion of the customer data from the data storage service according to the metadata.