How is it possible that this server NEVER loses data?

When you make as many videos as we do, you need a lot of fast, reliable storage. Our main editing server, Wanic, has checked all those boxes for years. It’s a great little server, built out of high-quality components and it even looks cool. But as our team has grown, we’ve reached the point where even a minute of downtime costs over $50. That’s just in payroll. So, practically speaking, the way to mitigate that is by adding redundancy.

Our drives are already redundant, with 20 drives in there with data striping. But the problem is they all sit in a single server. It’s been over a year in the making, but it’s finally here – WCK final form. I’m calling it Wanic 10 because it’s the last wever availability. Now, I’ve told you this like 10 times, nobody even knows what high availability means.

So, what does high availability mean? It means it’s robust. Just go ahead, unplug one, do it. Well, okay, I should probably tell you the stakes before you do that. Each of these two grand twin boxes has four entire servers inside of them, provided by Super Micro who sponsored this whole thing. They’re set up with WCA, a redundant NVMe first file system. In this config, it should sustain two entire servers dropping out without anyone even noticing.

Except that we moved the entire team onto it last night without telling anyone, and it’s the middle of the workday with a ton of high-priority videos in progress. Do you really want to test it right now?

A huge part of a project like this is the software, the stuff that’s going to handle distributing all of our terabytes of video projects, Word documents, and Linux ISOs to the multiple machines. But we can’t install any software until we have some hardware. So, let’s start there.

Meet the Super Micro Grand Twin A+ server, as-2115 GT-HNTR. Despite its sort of ordinary appearance and unexciting sounding name, it is anything but ordinary and very exciting. This 2U server has four independent computers inside, each with a completely independent motherboard, 384 gigs of memory, an AMD Epic Genoa processor with 64 cores, dual M.2 slots for redundant boot drives, six PCIe Gen 5 2 and 1/2 inch NVMe slots up front, and I/O in the rear.

The server slides into the chassis, which holds a modular backplane assembly and then passes through to the back of the server where you’ve got a management port, USB ports, display output, and power for the entire server. Each power supply is 2200 watts, 80 plus titanium, capable of handling four 400-watt Epic Genoa CPUs along with a bunch of RAM, NVMe drives, and network cards.

For our CPU, we’ve gone with an Epic Genoa 9534, a 64-core, 128-thread CPU with a max boost of 3.7 GHz, 4 gigabytes of level three cache, a 300-watt TDP, and support for DDR5 memory, up to 12 channels, and 128 lanes of PCIe Gen 5.

The memory is populated with 32-gigabyte DIMMs of DDR5 ECC running at 4800 megatransfers per second, totaling 384 terabytes of memory across all eight nodes. The boot drives are consumer-grade Sabrent 512-gig Gen 3 Rocket drives.

For storage, we’re installing two Kioxia Speedy CD6 Gen 4 NVMe drives in each node, one 7TB and another 15TB. These are placeholders for now, with plans to switch to 415TB drives per node.

Our network cards include ConnectX-6 200 Gbit cards from Nvidia, which terminate in OCP 3.0 small form factor mezzanine slots. Each node is connected to an Nvidia SN 3700 32-port 200 Gbit switch, running at non-maximum speed due to older Gen 4 cards.

The setup with redundant nodes allows for two servers to drop out without any noticeable impact, making it an ideal high-availability solution for our storage needs. The WCA software, specifically designed for NVMe drives, allocates specific CPU cores for dedicated tasks, ensuring low latency and high performance for Adobe Premiere editing and other data-intensive workloads.

With the capabilities of the Super Micro server and the WCA software, we can achieve speeds of over 1 terabyte per second, making our editing workflow faster and more efficient. The AI detection capabilities powered by Axle AI further enhance our video editing process by analyzing and tagging footage automatically.

Overall, the deployment of the Super Micro servers and WCA software has significantly improved our storage performance and high availability, allowing for seamless video editing and data management. The future expansion and optimization of our storage infrastructure promise even greater efficiency and productivity for our team.