More Than Capacity: Performance-oriented Evolution of Pangu in Alibaba

This paper presents how the Pangu storage system continuously evolves with hardware technologies and the business model to provide high-performance, reliable storage services with a 100-µs level of I/O latency. Pangu's evolution includes two phases. In the first phase, Pangu embraced the emergence of solid-state drive (SSD) storage and remote direct memory access (RDMA) network technologies by innovating its file system and designing a user-space storage operating system. As a result, Pangu substantially reduced its I/O latency while providing high throughput and IOPS. In the second phase, Pangu evolved from a volume-oriented storage provider to a performance-oriented one. To adapt to this business model change, Pangu upgraded its infrastructure with storage servers of much higher SSD volume and RDMA bandwidth from 25 Gbps to 100 Gbps. It introduced a series of key designs, including traffic amplification reduction, remote direct cache access, and CPU computation offloading, to ensure Pangu fully harvests the performance improvement brought by hardware upgrades. Other than technology innovations, we also shared our operating experiences during Pangu's evolution, and discussed important lessons learned from them.