US 12,222,908 B2
Detecting idle periods at network endpoints for management actions at processing clusters for managed databases
Ippokratis Pandis, Menlo Park, WA (US); Eric Ray Hotinger, Redmond, WA (US); Bruce William McGaughy, Mercer Island, WA (US); Naresh Chainani, Mountain View, CA (US); Neeraja Rentachintala, San Jose, CA (US); Zhixing Ma, Kirkland, WA (US); Pulkit Jagdishchandra Bhavsar, The Ponds (AU); Chao Duan, Seattle, WA (US); William Michael McCreedy, Berlin (DE); Pavel Sokolov, Menlo Park, CA (US); and Sanjay Wangoo, San Ramon, CA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Nov. 26, 2021, as Appl. No. 17/535,940.
Prior Publication US 2023/0169048 A1, Jun. 1, 2023
Int. Cl. G06F 16/21 (2019.01); G06F 16/23 (2019.01)
CPC G06F 16/21 (2019.01) [G06F 16/2358 (2019.01); G06F 16/2393 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
one or more computing devices, respectively comprising a processor and a memory that implement a processing cluster for a database service, the processing cluster comprising a leader node and one or more compute nodes, wherein the leader node implements query planning and execution to instruct the one or more compute nodes in accordance with a query plan generated by the leader node, and wherein the leader node in the processing cluster is configured to:
monitor information indicative of client activity received via a network endpoint at a proxy service associated with a database managed by the database service to detect an idle period for the database, wherein the processing cluster including the leader node receives client requests from the proxy service that are directed to the database over a network;
select from a plurality of different management actions a management action for the processing cluster to perform during the detected idle period for the database according to the detected idle period; and
send one or more requests to cause the management action for the processing cluster to be performed, wherein the database is available for read requests received via the network endpoint during performance of the management action by another processing cluster of the database service.