Year: 2021
Type: conference-paper
Abstract: RocksDB is a general-purpose embedded key-value store used in multiple different settings. Its versatility comes at the cost of complex tuning configurations. This paper investigates maximizing the throughput of RocksDB 10 operations by auto-tuning ten parameters of varying ranges. Off-the-shelf optimizers struggle with high-dimensional problem spaces and require a large number of training samples.
Authors: Sami Alabed, Eiko Yoneki
Institution: University of Cambridge
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Funder: Alan Turing Institute
OpenAlex ID: W3139820007