网页 综合 更新时间 2026-11-16 21:45:15

Managed Network File Storage for AI workloads | DigitalOcean

自动蜘蛛 www.digitalocean.com 正文约 2747 字
Managed Network File Storage for AI workloads | DigitalOcean
Featured AI Products
Build, deploy, and scale cloud compute resources
Containers and Images
Safely store and manage containers and backups
Fully managed resources running popular database engines
Management and Dev Tools
Control infrastructure and gather insights
Secure and control traffic to apps
Help protect your account and resources with these security features
Store and access any amount of data reliably in the cloud
Security and Networking
Web and App Platforms
Store and scale file storage for AI/ML and data-intensive workloads with fully managed high-performance service without the operational burden.
Create and manage POSIX-compliant NFSv3 and NFSv4 shares in your virtual private cloud with fully managed service.
Mount NFS shares to multiple GPU, CPU Droplets and Kubernetes (DOKS) clusters to share a dataset across AI/ML training, inference, and containerized applications.
Optimize high-throughput, low-latency storage for AI/ML training and inference workloads.
Create and restore point-in-time snapshots of NFS shares to quickly restore data.
See Regional availability.
Standard tier starts at $0.15 per GiB per month with aggregated throughput of up to 1 Gbps per share. Start with 50 GiB.
High performance tier starts at $0.30/GiB/month. Throughput starting from 8 Gbps. Start with 500 GiB.
Snapshots billed separately at $0.06 per GiB/month based on restorable file size.
Flat pricing across all supported regions.
Network file storage (NFS) is a fully managed, POSIX-compliant file storage service that allows multiple servers or containers to access the same shared data over a private network. It is ideal for AI/ML training, inference, and other data-intensive workloads.
Network File Storage is available in Atlanta, New York and Amsterdam data centers, with more regions planned in future expansions.
NFS shares can be mounted on GPU Droplets optimized for AI/ML workload and DigitalOcean Kubernetes (DOKS) clusters via statically provisioned PersistentVolumes (PV) and PersistentVolumeClaims (PVC). This makes it easy to share datasets across multiple nodes.
Pricing of NFS is allocation based. You pay a flat monthly fee per provisioned GiB, with no additional charges for operations or access requests. Standard tier starts at $0.15 per GiB per month with throughput of up to 1 Gbps per share. High performance GPU tier starts at $0.30 per GiB per month with discounts available for GPU-committed customers. Snapshots billed separately at $0.06 per GiB/month based on restorable file size.
Learn more from our product documentation.
From GPU-powered inference and Kubernetes to managed databases and storage, get everything you need to build, scale, and deploy intelligent applications.
© 2026 DigitalOcean, LLC.
访问原文 返回搜索 https://www.digitalocean.com/products/storage/network-file-storage
本页为82搜索生成的网页快照,内容来自站内索引;演示数据的原文链接仅作示例。