Turbocharging Cloud-Native Storage for Database Workloads
2020-03-29, 14:05–14:35, Room 2

Rook is a cloud-native,open-source storage orchestrator that solves the problem of storage deployment and management in kubernetes. But it still does not solve the problem of getting the most out of your storage for databases. We will be able to unlock the potential of the cloud-native storage for database workloads through performance analysis and tuning. Quantifying storage performance in general is problematic, as it requires an understanding of using trusted benchmarking tools, and running these benchmarks on their own can be challenging. This gets further complicated in cloud native environments, where the access to the storage backend is abstracted. To address this, We will be introducing methodologies and tools to demystify the performance of cloud native storage. We will also be discussing a case study of tuning a rook deployment for postgres, this will assist the audience with resources to gather performance data of cloud-native storage and then using the data to tune the storage backend for databases.