Answer to: Recommended data structures/algorithms for checking peoples' availability schedules

Doing search service side is a very bad idea. You would have to keep assignments synchronized with DB doing error-prone cache invalidation and potentially moving significant volumes of data around. Do not do full scans service-side. Use DB to do searches: CREATE TABLE workers ( name VARCHAR NOT NULL PRIMARY KEY ) CREATE TABLE assignments ( worker VARCHAR NOT NULL REFERENCES workers(name), during tsrange NOT NULL, CONSTRAINT overlapping_times EXCLUDE USING GIST ( worker WITH =, during WITH && ) ) INSERT INTO workers VALUES ('John'), ('Mary') INSERT INTO assignments VALUES ('John', '[2021-01-01 00:00:01, 2021-01-01 00:09:01)'), ('John', '[2021-01-01 00:13:01, 2021-01-01 00:14:01)'), ('Mary', '[2021-01-01 00:05:01, 2021-01-01 00:16:01)') Find busy workers: SELECT worker from assignments WHERE during && tsrange('[2021-01-01 00:10:01, 2021-01-01 00:11:01)') worker Mary Find available workers: SELECT name FROM workers LEFT JOIN assignments ON assignments.worker = workers.name AND during && tsrange('[2021-01-01 00:10:01, 2021-01-01 00:11:01)') WHERE assignments.worker IS NULL name John fiddle I'm not sure which additional indexes would be optimal for this case, but btree-gist module surely has necessary operators to build them. There is a chance, that all indexes that might be needed are already created by CONSTRAINT directives.