Distributed Data Placement with NVMe-oF - Design and Implementation Challenges

Explore a 36-minute conference talk from SNIA Storage Developer Conference 2024 focusing on distributed data placement algorithms in NVMe-oF environments. Dive into Simplyblock's containerized distributed lock storage system, optimized for AWS environments and designed to maximize local NVMe IOPS with network latency under 200 microseconds. Learn how virtual volumes achieve high availability through erasure coding and self-balancing cluster capabilities. Examine common challenges in distributed architecture for reliable storage scale-out, particularly in high-performance contexts using NVMf-TCP and NVMf multipathing. Master fundamental design principles for scalable data placement algorithms in cloud storage, understand the complexities of NVMf-based environment scaling, and grasp essential concepts of distributed erasure coding implementation. Presented by Michael Schmidt from simplyblock.io, gain practical insights into building robust, high-performance distributed storage systems.