On March 17, 2025, Google Cloud launched a hierarchical namespace (HNS) feature in Cloud Storage to optimize AI and machine learning (ML) workloads. HNS allows atomic folder-level operations, which improve data organization and reliability. Compared to traditional flat namespaces, HNS enables quicker checkpointing, with benchmark tests showing up to a 20-times acceleration in checkpoint writes. Real-world applications like AssemblyAI reported substantial performance improvements, including a 10x increase in throughput and a 15x improvement in training speed. HNS enhances query performance, making it ideal for large-scale AI/ML processes.
Google Cloud Storage (GCS) Hierarchical Namespace (HNS) accelerates storage workloads that rely on filesystem semantics-like folder renames-boosting performance significantly, especially for AI/ML tasks.
AssemblyAI reported a 10x increase in throughput from Google Cloud Storage when using HNS with Cloud Storage FUSE, leading to a 15x improvement in training speed.
Collection
[
|
...
]