IBM Spectrum Scale is an enterprise Software Defined Storage product which means it is scalable, high performance, full featured, and never seen in the home environment. But since last year there is a special version of Spectrum Scale which is available free for home or development use and contains all features, but is capped at 12 TB of storage. Compared to the building the Summit Supercomputer that has 250 PB of Spectrum Scale storage with 4608 nodes, a home NAS should be easy right, I mean, how hard can it be?
Another question is, why would you want to run an enterprise product like Spectrum Scale at home? Just use a Synology, QNAP, or some dedicated Linux distribution like FreeNAS. But where is the fun in that? If you want total control over where your data is stored, and use enterprise level management and cloud access features, this is your chance. You'll have to know command line, and have a little Linux experience, but it's all very doable, really.
In this blog series you can follow me while I install Spectrum Scale on a small Intel computer, configure file sharing, migrate some data, create snapshots, and move files to the cloud. You can do the same on a Virtual Machine on your laptop, I'm only using two cores and 4GB of memory to do all this.
These are the blog parts:Part 1: Installing and Creating the ClusterPart 2: Creating a file sharePart 3: Extending the NASPart 4: Migrating files to the best storagePart 5: Creating snapshots to safeguard against ransomwarePart 6: Backing up to the IBM Cloudpart 7: Maintenance and Trouble Shootingpart 8: Tuning and Quotas
To give you a taste of what the result looks like, here's a command line example:
GPFS cluster information
GPFS cluster name: scalenode1
GPFS cluster id: 17518623129147820244
GPFS UID domain: scalenode1
Remote shell command: /usr/bin/ssh
Remote file copy command: /usr/bin/scp
Repository type: CCR
Node Daemon node name IP address Admin node name Designation
1 scalenode1 192.168.178.199 scalenode1 quorum-manager-gateway-perfmon
There is a GUI available as well for management and monitoring tasks:#Highlights#Highlights-home