What is EdgeFS?
EdgeFS enables high throughput and low latency access for stateful applications via standard storage protocols (NFS, S3 and iSCSI). Outstanding performance characteristics for variety of use cases: HDD-only, SSD/HDD hybrid and SSD/NVMe only.
EdgeFS spans unlimited number of geographically distributed sites, connected as one global Geo-Transparent namespace. It scales equally well for Object, Block or File access within local site and enables efficient access to data on remote sites.
EdgeFS can be installed in any cloud on top of existing or new Kubernetes cluster. It integrates with Kubernetes CSI and Rook orchestration to provide easy administration, unlimited file, object, directory or bucket level granularity snapshots.
Deep Kubernetes Integration
With Rook running in the Kubernetes cluster, Kubernetes PODs or External applications can mount block devices and filesystems managed by Rook, or can use the S3/S3X API for object storage.
The Rook operator is a simple container that has all that is needed to bootstrap and monitor the storage cluster. The operator manages CRDs for Targets, Scale-out File (NFS), Object stores (S3/S3X), and Scale-out Block (iSCSI) by initializing the pods and other artifacts necessary to run the services.
The EdgeFS operator monitors storage Targets to ensure that cluster is healthy. EdgeFS will dynamically handle services, devices, servers or site-zones failover.
The EdgeFS Rook operator comes with tightly integrated CSI plugin for static and dynamic Persistent Volume provisioning via NFS or iSCSI.
Rook is implemented in golang. EdgeFS is implemented in Go/C where the data path is highly optimized.
File, Block and Object
Supports variety of storage protocols on top of same geo-distributed namespace - S3, S3X (Range Writes, Snapshots, etc), NFS, iSCSI
Fast & easy deployment
Deep Rook.io and CSI integration. Deploy in minutes on AWS, Azure, GCP, On-Prem or Edge IoT.
Data placement is replicated and offline erasure encoded. Fault-tolerant with built-in self healing and advanced snapshotting capabilities.
Geographically-aware de-duplication and on the fly compression. Reduces cross-site link utilization and noticeably speeds up remote data access.
At rest encryption
Strong server-side in-software encryption for selected files, LUNs, buckets or objects. All data and metadata immutable and crypthografically self-validated with SHA-3 256-bits or 512-bits.
Multi-Tenancy with QoS controls
Geographically-aware multi-tenancy with file, LUN or object granularity level QoS feature set.
No single point of failure (SPoF) due to fully distributed, immutable metadata design and no need for dedicated metadata servers.
Architected with microsecond resolution I/O failover guarantees utilizing location-independent data placement and retrieval technique.