Fur Affinity prepares day-long move to new monster hosting
Furry art site Fur Affinity has announced its migration to new hosting on Wednesday, September 9. Previously set for August, then Monday, the final migration "may take" 18 hours; during which content modification will be disabled, but interactions will work.
The new hardware was announced in April: a "cluster-based server platform" named after a kaiju with 112 CPU cores, 1536GB of RAM and 153.6TB of "enterprise SSD".
If composed of four 28-core Xeon CPUs, 24x 6.4TB SSDs and 48x 32GB DDR4-2666 ECC DIMMs, the components alone have a retail value of ~US$100,000 – not including servers or switches.
While databases are commonly replicated natively between machines, allowing for fast switchover, content is typically synchronized by tools such as rsync requiring a static archive for consistency.
Aside from "chonkier bandwidth", and the much-anticipated potential for larger uploads, FA staff were upbeat about the new platform:
This upgrade will be one of the largest, most substantial upgrades in the site's history. It will provide faster loading times, make it easier for us to provide new services and features as well as allowing us to implement some quality-of-life changes to help improve the user experience.
Fur Affinity Forums, missing in action since a power outage on July 22, are expected to return "a few days after" in an upgraded form.
Update (Sep 9): Uploads are now available again; the process took 22 hours.
About the author
GreenReaper (Laurence Parry) — read stories — contact (login required)a developer, editor and Kai Norn from London, United Kingdom, interested in wikis and computers
Small fuzzy creature who likes cheese & carrots. Founder of WikiFur, lead admin of Inkbunny, and Editor-in-Chief of Flayrah.
Comments
Aw, crap!
To put it into perspective: this isn't a ridiculous amount of storage, but even in 2020 I'm mildly surprised they went all-in on SSD. Tiered storage tends to be more cost-effective. Inkbunny's main server has a 500GB NVMe SSD and 4x4GB HDDs: database and thumbnails reside on the SSD (read latency: ~0.45ms), while submissions are on HDD (~7.4ms). A big difference, but data accessed frequently typically remains in RAM. Let's hope 1.5TB is enough for FA! It's unclear what they meant by a "cluster-based server platform", but I'm inclined towards a 'chonky' box like the ThinkSystem SR950, given that a "three-headed hydra" (application server, database and storage?) shares one body.
There are other reasons to go all-in on SSD, such as power usage and reliability; perhaps those are a bigger concern where it's going. Or maybe they expect to read from storage more frequently – I recall FA's thumbnailing system acts as a cache rather than making immutable copies; scaling tends to be a high-CPU operation, so may drive core-count. Databases can parallelise to many cores, too.
They'll presumably still have their old hardware, which might be put out to pasture for backup storage or secondary processing. It might not stay in North Virginia, though – I guess they'll end up close to Redwood City, where IMVU is based. Judging by this, their datacenter is in Oakland; but they've been migrating to AWS. So maybe it's not a box at all, but Kurbernetes in the cloud? But then, they wouldn't be talking about servers and switches.
My first impression is that yes, tiered storage is generally more cost-effective.
Toho will probably not protect their IP, but I'm not happy they officially call the server King Ghidorah. Kaijus are a Japanese establishment and they're doing them a disservice by having a copyrighted one relate to the Fur Affinity enterprise.
If I had a couple millions lying around I would beat the shit out of this fake King Ghidorah with the real King Ghidorah, like Toho did with the American Godzilla back in Godzilla: Final Wars (2004).
You're right; it might technically violate Toho's "King Ghidorah" trademark, which includes the class of computer equipment.
That said, I doubt Toho is bothered by this use, unless they plan on bringing out kaiju-brand servers (which'd be awesome). Besides, they'd probably want to go after Corsair first - they're using it to push hardware sales, not a $5 site subscription.
At the end of the day, storage caused FA issues in the past. Hard to blame them for throwing $40k to make such issues go away.
I'd say Corsair is a more respectable brand.
Post new comment