Skip to content

Distributed File Systems

September 29, 2023
May 20, 2015

enterprise-server#Storage

Clustered file system - Wikiwand
List of Distributed file systems - Wikiwand
Comparison of distributed file systems - Wikiwand

File systems unfit as distributed storage backends: lessons from ten years of Ceph evolution – the morning paper

n1trux/awesome-sysadmin: Distributed Filesystems

Distributed Storage: Picking The Right Tool For The Job | SolidFire | Blog
Distributed Storage Tech Preview [with video] - VMware vSphere Blog - VMware Blogs (2012)
File storage, block storage, or object storage?
Distributed File System SBFAQ

Distributed Replicated Block Device - Wikiwand

filesystems - Distributed File Systems: GridFS vs. GlusterFS vs Ceph vs HekaFS Benchmarks - Stack Overflow
Ceph at CERN: A Year in the Life of a Petabyte-Scale Block Storage Service » OpenStack Open Source Cloud Computing Software

Testing of several distributed file-systems (HDFS, Ceph and GlusterFS) for supporting the HEP experiments analysis - IOPscience PDF
Performance comparison of distributed file systems, Marian Marinov (1H Ltd) on Vimeo
[Linux.conf.au 2013] - grand distributed storage debate glusterfs and ceph going head head - YouTube
Ceph at CERN: A Year in the Life of a Petabyte-Scale Block Storage Service - YouTube

Performance comparison of Distributed File Systems on 1Gbit networks

Global File System 2 - Red Hat Customer Portal
Hortonworks Data Platform -Hortonworks Data Platform

BeeGFS

BeeGFS – The Leading Parallel Cluster File System

Ceph

Ceph (software) - Wikiwand
ceph.com
Ceph is a flexible object storage system, with four access methods: Amazon S3 RESTful API, CephFS, Rados Block Device and iSCSI gateway.

Note that ceph has several aspects: rados is the underlying object-storage, quite solid and libraries for most languages; radosgw is an S3/Swift compatible system; rbd is a shared-block-storage (similar to iSCSI, supported by KVM, OpenStack, and others); CephFS is the POSIX-compliant mountable filesystem.

5 Ceph storage questions answered and explained
GlusterFS vs. Ceph: Weighing the open source combatants The foundation of Ceph is object storage, that of GlusterFS is a file system.

Red Hat Ceph Storage | Red Hat

Bootstrap your Ceph cluster in Docker
ceph/ceph-docker: Docker files and images to run Ceph in containers

use Rook

Ceph Intro and Architectural Overview by Ross Turk - YouTube
Ceph Intro & Architectural Overview - YouTube
Fundamentals of Ceph by Greg Farnum - YouTube
Storage Tutorial - Learning Ceph - YouTube

Rook

Rook.io
File, Block, and Object Storage Services for your Cloud-Native Environments
Kubernetes Operator for Ceph (and many others)

Rook’s Framework for Cloud-Native Storage Orchestration
Why you should master Rook for Ceph storage on Kubernetes - Superuser
Simplify Storage for Kubernetes with Rook and Ceph - The New Stack

To Rook, or not to Rook, that’s Kubernetes - Flant - Medium
Manual recovery of a Rook cluster in Kubernetes - Flant - Medium

Ceph storage with Rook Running Ceph on Kubernetes - YouTube
KubeCon Barcelona: Rook, Ceph, and ARM: A Caffeinated Tutorial | Sébastien Han
Storage 101 Rook and Ceph - YouTube
YVR18-114:Auto-deployment of Ceph cluster with Rook on top of Kubernetes - YouTube Strong accent
Auto-Deployment of Ceph Cluster With Rook on Top of Kubernetes - Dennis Chen, Arm - YouTube English interpretation

EdgeFS

EdgeFS
Kubernetes Rook EdgeFS 1.1 Released - ITNEXT

GlusterFS

GlusterFS - Wikiwand
Storage for your Cloud. — Gluster
Gluster Docs
Architecture - Gluster Docs

Red Hat Gluster Storage (formerly Red Hat Storage Server) | Red Hat
Red Hat Gluster Storage architecture | Red Hat

LAMP Cluster — Distributed filesystem • Websites, Hosting and Friends
GlusterFS performance on different frameworks • Websites, Hosting and Friends

Demystifying Gluster - GlusterFS For SysAdmins - YouTube

GlusterFS – JamesCoyle.net

Block Storage

LINSTOR/DRDB

LINSTOR | LINBIT-Creates, Removes, & Manages storage volumes
Distributed Replicated Block Device - Wikiwand
DRBD - Linux-HA

Linbit Docs – Docs LINBIT

iSCSI

Most NAS can expose storage pool as iSCSI LUN over IP.

iSCSI Target: An iSCSI storage server. In this tutorial the target is your NAS.
iSCSI initiator: An iSCSI client. Initiators connect to targets and use their storage.
iSCSI LUN: a portion of storage space that can be utilized by initiators by connecting it to a target. LUNs can be block-based or file-based, though block-based is recommended as it supports more features.

Warning: Connecting more than one initiator to the same target might result in data loss or damage to the NAS disks.

How to create and use the iSCSI target service on a QNAP NAS - QNAP (AU)

Commercial Solution

HPE Alletra 9000
关于玩转高端存储的实景教学:环境搭建与配置 - YouTube

Object Storage

aws#S3

kahing/goofys: a high-performance, POSIX-ish Amazon S3 file system written in Go
s3fs-fuse/s3fs-fuse: FUSE-based file system backed by Amazon S3

Minio

MinIO | High Performance, Kubernetes Native Object Storage self-hosted S3-compatible object storage
Minio Docs

The complete guide to attach a Docker volume with Minio on your Docker Swarm Cluster
Kubernetes object storage best practices | InfoWorld

OrangeFS

OrangeFS - Wikiwand
OrangeFS more info than orangefs.org

An open source FS by EMC, the one behind
EMC Isilon - Wikiwand

Stratis

Stratis Storage

Stratis: Easy local storage management for Linux [LWN.net]

Lustre File system

Lustre (file system) - Wikiwand
Lustre

The Lustre Distributed Filesystem | Linux Journal
Lustre® File System | OpenSFS: The Lustre File System Community

White papers:
Inside The Lustre File System by Seagate
Lustre File System bu Sum

Lustre File System - YouTube

XtreemFS

XtreemFS - Wikiwand
XtreemFS - Fault-Tolerant Distributed File System

I tested XtreemFS and found that it does not work well. There are problems like data corruption (#359), read errors in degraded mode (#357/#235), crippled read-only mode (#358) etc.; build system is a mess plus XtreemFS depends on old (not updated since 2007) non-free JAR (#309, #173) so XtreemFS is in violation of DFSG and not distributable in Debian. Also I'm not happy about how devs respond to bugs. Finally XtreemFS is written in poor language notorious for inefficient memory management so naturally XtreemFS can't stand against GfarmFS and LizardFS in performance comparison.

IPFS

IPFS is the Distributed Web
How To Get Started With IPFS and Node - Better Programming - Medium
Guide to IPFS garbage collection - LogRocket Blog
The IPFS Protocol Explained with Examples - Welcome to the Decentralized Web - YouTube
Measuring the Web3.0 Stack - YouTube

Cloudflare
Serving Cloudflare Pages sites to the IPFS network Primer
Gaining visibility in IPFS systems InterPlanetary Naming System (IPNS), DNSLink
Distributed Web Gateway | Cloudflare
cloudflare/ipfs-gateway-monitor

Pinata | Your Home for NFT Media

IPFS-based database
Home – OrbitDB
ThreadDB | Documentation | Textile

FileCoin

A decentralized storage network for humanity's most important information | Filecoin
Storage providers earns a fee

HDFS

LizardFS

LizardFS

SeaweedFS

chrislusf/seaweedfs: SeaweedFS is a simple and highly scalable distributed file system.

OpenEBS

OpenEBS block storage
Welcome to OpenEBS Documentation ·

OpenEBS · GitHub
Using OpenEBS as a Kubernetes persistent volume – openebs

Longhorn

Longhorn
longhorn/longhorn: Cloud-Native distributed storage built on and for Kubernetes
longhorn/longhorn-manager: Millions and millions of volumes orchestrated

Project Longhorn Now Available on Kubernetes
Longhorn: Rancher's Journey from Zero to GA
Longhorn: A Cloud Native Approach to Distributed Storage - YouTube

Cloud Native Distributed Storage in Kubernetes with Longhorn - YouTube