Efficient Key-Value store for future hardware

The objective of the project is to design an efficient key-value store for future hardware, using fast NVMe SSDs and persistent byte-addressable memory.

In our modern digital life, activities from buying clothes to accessing government services depend on working with computer applications which store information in a long-lasting form, that is, they need a data store, where information (called ”values”) is found using a label or ”key”, So the key-value store is vital for the functionality and performance of all applications. Even relational databases typically contain a key-value store as a storage engine, underneath layers that support a richer query model.

The current key-value stores are designed for today’s typical hardware environment with a memory hierarchy of slow but capacious hard disk, flash-based SSD, and fast but limited and volatile RAM. Currently, no key-value store can leverage the speed of fast NVMe SSDs and byte-addressable persistent memory -- existing key-value stores become CPU bound before reaching the limit of these devices.

The goal of this project is to rethink the way data is stored in memory and on disk to design a fast key-value store for modern drives and persistent memory. The project will start with a study of the characteristics of persistent memory. Then existing data-structures designed for volatile DRAM will be ported and optimized for persistent memory. Finally, new designs will be proposed.

Requirements: fluent C, C++, or Rust programming, notions of good practices in optimisations and efficient programming.

Scholarships value and duration: $35.5k for 3.5 years

Enquiries:

Dr Baptiste Lepers: [Email Address Removed]