These benchmarks exercise storage apis in a real-life usage way (avoiding microbenchmarks).
This models an offline load of a Google doc. See this document for a breakdown of the database and the transactions, along with the traces used to extract this information.
The cold open variant of this test opens the database from disk in each iteration to more realistically model activity on a fresh user navigation.
Models different scenarios pertaining to a mail client that caches data in IndexedDB for offline use.
This benchmark models the creation and reading of blobs. It has two parts:
There is a variant of this test for every transportation type (shared memory, files, and ipc).