CARBON-AWARE DISTRIBUTED QUERY SCHEDULING IN CLOUD-NATIVE DATABASES

Abstract

Data centers account for a growing share of global electricity consumption, generating substantial carbon footprints due to reliance on fossil-fuel-powered energy grids. While cloud-native distributed databases (such as CockroachDB and Amazon Redshift) optimize analytical query scheduling for monetary cost, execution speed, and latency SLAs, they typically ignore real-time variations in grid carbon intensity ($g\text{CO}_2\text{eq/kWh}$). This paper introduces a Carbon-Aware Query Scheduler (CAQS) for cloud-native distributed analytical databases. By combining real-time carbon intensity telemetry with dynamic query routing, regional migration, and deferral heuristics, CAQS reduces operational carbon emissions while adhering to strict execution Service Level Agreements (SLAs).