SNIA SDC 2025  - Discussion & Analysis of the MLPerf Storage Benchmark Suite & AI Storage Workloads @SNIAVideo
SNIA SDC 2025  - Discussion & Analysis of the MLPerf Storage Benchmark Suite & AI Storage Workloads  @SNIAVideo
Uploaded November 2025 | Updated November 2026, 1 day ago
Storage for AI is rapidly changing: Checkpointing becomes more important as clusters scale to more accelerators, managing large KV-Caches from LLM queries shifts inference bottlenecks to storage, accessing relevant data via VectorDB similarity searches drives small IOs for nearly every query, and future applications may require wildly different storage architectures. The MLPerf Storage v2.0 Benchmark Results were just released and the v2.5 suite is under active development. In this session we go over the Checkpointing benchmark added to v2.0 and the KV Cache Management and VectorDB benchmarks planned for v2.5. We will analyze IO traces of the workloads to understand how these workloads translate from the application layer to storage and discuss how these workloads may drive future storage performance requirements.

Storage requirements for AI data centers are shifting quickly as the infrastructure to support AI systems needs to support the exponential increases in AI performance. The MLPerf Storage working group develops a benchmark suite of representative applications that can ease the process of understanding storage for AI applications. Future storage systems may need drastic changes to continue to match the pace of improvements seen from the AI software and hardware.
Presented by Wesley Vaske, Micron Technology
Learn More:
• SDC Website: snia.org/sniadeveloper
• SNIA Website: snia.org
• SNIA Educational Library: snia.org/library
• X: twitter.com/SNIA
• LinkedIn: linkedin.com/company/snia
SNIA SDC 2025  - Discussion & Analysis of the MLPerf Storage Benchmark Suite & AI Storage WorkloadsPerformance-based TCO Model 2.0 WalkthroughSNIA SFF TWG - 2026 SNIA PreviewLiquid Cooling, SSD Design, And Standards For The AI EraSNIA Cloud Storage Technologies Community - 2026 SNIA PreviewSNIA Partner Update - TGG - June 2024SNIA Member Recognition Program - 2026 SNIA PreviewSNIA SDC: StorageAI 2026 - NVMe LBA Access Control for GPU-Direct Storage in AI/HPC WorkloadsCombinatorial 4-way and 6-way Strand Displacement for Programmable Reversible Data ManipulationRobust DNA Data Storage via Block-by-Block Synthesis: The PEARL DNA ArchitectureAI Stack: From Data to Decisions: Understanding How AI Models LearnSNIA Green Storage Community - 2026 SNIA Preview
SNIAVideo |

SNIA SDC 2025 - Discussion & Analysis of the MLPerf Storage Benchmark Suite & AI Storage Workloads

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER