David Wang | Director, Enterprise SSD Firmware
Silicon Motion Inc.

Future of Memory and Storage - Day 2 @ 09:05

Fine-Grain I/O Acceleration in SSD Controllers for Agentic AI Workloads

Agentic AI workloads increasingly require near-GPU storage access with ultra-fine-grain I/O, often well below the conventional 4 KiB block size. Emerging GPU-initiated storage models highlight the need to sustain very high IOPS while efficiently utilizing PCIe Gen6/Gen7 bandwidth, potentially with fewer SSDs per GPU domain.However, mainstream data-center SSDs are architected around 4 KiB host I/O assumptions, targeting millions of IOPS per device at higher PCIe generations. When host I/O sizes shrink to 512 B or smaller, overall throughput degrades sharply due to PCIe command and completion overheads, interrupt pressure, and limited per-I/O processing capacity within the SSD controller. As a result, achieving even a modest fraction of peak PCIe bandwidth under fine-grain I/O becomes a fundamental architectural challenge.This presentation introduces a set of multi-level I/O coalescing techniques in next-generation SSD controllers designed to address these bottlenecks. The proposed approach spans NVMe queue-level aggregation and command-level coalescing, substantially reducing PCIe transaction overhead per effective I/O while preserving host-visible fine-grain I/O semantics.

David Wang, Director, Enterprise SSD Firmware, Silicon Motion Inc.

×

David Wang / Fine-Grain I/O Acceleration in SSD Controllers for Agentic AI Workloads

David Wang, Director, Enterprise SSD Firmware,
Silicon Motion Inc.

David Wang is a Technical Director at Silicon Motion, where he has led HW/FW co-design and FW architecture design for PCIe Gen5, Gen6 and next-gen enterprise SSD controllers since 2020. His work focuses on high-performance I/O processing, controller micro-architecture, optimized HW/FW codesign for...

CloseMore