Live backup PC
Live backup PC
That's a good idea. I'll have to check the software licensing though, as there are a lot of restrictions on that. At my price point ($50k), each license is node-locked to a physical PC by MAC address. They also offer network/floating licenses, I did that once several years ago when I was operating an environment with several users, but the floating licenses run about $75k per seat. Since I'm now a single user and self-funded, a spare $2k refurb PC looks pretty attractive against that pricing.
I don't need more or faster storage. I need a spare PC, for when the power supply or CPU or motherboard fails on the one I'm using.
Wow. I'll definitely have to talk with you, next time I'm looking to replace my primary system. I usually just buy what the software manufacturer recommends, as they do a lot of benchmark testing specifically with their software. Right now, I'm running a dual Xeon Gold 6226R, which of course has a lower clock speed than some of the options available when I purchased (mid-2022), but actually benchmarked better than some of the systems running higher clock speeds.
But right now, I'm not looking to replace this system, just looking to set up a cheap spare to handle failures.
This makes perfect sense. For your approach of avoiding the need to pull backups from a NAS, it might be beneficial to discuss options with a company like Veeam. They can replicate virtual machines and provide physical device backups as well, but they may also support transferring data between workstations. Additionally, for VM replication, they offer continuous data protection which would significantly reduce downtime. You'd need to install switches so workstations can communicate, and for this level of replication you'd require at least 10GbE.
I'll definitely look into Veeam. I don't really require real-time replication. A delay of several minutes or even hours would suffice, and overnight replication would be acceptable, reducing the need for high Ethernet speeds.
Clearly, the tighter the timeframe, the less damage when a machine fails. I'm willing to accept losing an hour's work or even a day if necessary. However, sometimes errors occur—people forget about changes made yesterday and assume the backup they're using has fixes within that period.
That's frustrating. Given how costly this software is, it seems like they're asking users to give up something valuable, promising to retire the old machine before installing the new one, and not attempting to run multiple licenses simultaneously. They do have systems in place to detect and prevent this when connected to the internet, but they also know there are ways around them for those who aren't fully honest.
I currently have two software packages that need re-licensing each time this happens, though they usually resolve it within a day. During planned system upgrades every 2 or 4 years, it's not a major issue. But my last three Dell Precision workstations failed shortly after installation, which adds an extra layer of difficulty.
Side note: why does Dell now take three attempts to build a system that lasts? I've been using this kind of software for nearly 25 years, but the number of new Dell Precisions failing right out of the box dropped from almost zero before 2014 to nearly all in the last decade. I don't recall buying a system recently that didn't require a motherboard or power supply replacement within the first month. After that, they seem to have a much longer and more stable lifespan, which is the only reason I haven't completely given up on Dell, though they still have significant issues along the way.
I contributed to high-performance computing projects for a defense company. My role involved buying a large amount of hardware and software. I believe you should review your business strategy and assess whether floating licenses make sense given the scope of your tasks.
You might also want to explore RAM disks for better efficiency. With modern systems, TBs of RAM are common. The 6000 series Xeons feature 8-channel memory controllers. To reach peak performance, a dual-socket setup with 16 DIMMs is ideal. A rack full of sixteen 64GB DIMMs represents a significant amount of memory. This should be considered.
Additionally, testing your workflows with hyperthreading both enabled and disabled could reveal improvements. You may discover that performance drops without hyperthreading are better.
Some valuable perspectives from someone experienced in HPC environments.