The PanFS Data Engine Architecture
Used in manufacturing, life sciences, energy, research, financial services, semiconductor design and media & entertainment, the Panasas® PanFS® is the data storage solution of choice for unlimited scalability and performance, enterprise reliability, and ease of use. An integrated portable storage platform that combines a high-performance distributed and clustered parallel file system, volume manager, and software-based erasure-coding data protection, PanFS delivers datasets of the size and complexity required in scientific computing, modeling and simulations, high-performance data analytics, and AI/ML workloads. Built on a foundation laid over 20 years ago as the first parallel file system for Linux and now in its ninth generation of commercial deployment, the newly released PanFS 9.2 powers a portfolio of ActiveStor all-NVMe flash, hybrid and massive capacity data storage appliances.
We invite you to download the PanFS 9: Architectural Overview white paper, to learn more about PanFS including:
- Details of its control and data planes
- DirectFlow client driver
- Full POSIX semantics with cache coherency, data management and security, erasure coding
- Prevention of hot spots, single-tier solution, and patented technology for per-file object RAID