| |
| ▲ | 0manrho 2 days ago | parent [-] | | Correct, it also misses the core point of the discussion: SSD's and HDD's are unreliable cold storage long term. Cheap HDD's are better than Cheap SSD's (debatable if you're willing to spring for high end parts, but that's outside the scope of value/affordability), but if that data is truly important, it's well established best practice to replicate (be it cloning, mirroring, parity, what have you) your storage media. Sometimes, unfortunately, a user will value the data more than they can afford to properly replicate/secure it it, and compromises must be made. Also, Tape does require a higher buy in than most (individual) SSD's/HDD's demand before you can even start investing in actual storage media, even if going used, so there are absolutely valid contexts where the "right"/best available approach is just throwing $200 at second HDD and cloning/mirroring it, but best available compromise in any given specific context is a separate discussion from general best practices or best value in broad terms. |
|
| Correct, and absolutely worth noting, but point still stands. Had no intention of misleading, I called it a 6TB drive because that's what they're called (technically 6.25TB if we really want to get pedantic). Whether using LTO's compression or not, whether your data is already compressed or not, it's still a reasonable affordable, dense, reliable, approachable cold storage offering. Same is true even for LTO5. It only starts to go sideways when you step up to LTO7 and above or try to get an autoloading all-in-one library unit. Though you can get lucky if you're patient/persistent in your bargain hunting. |
| |
| ▲ | stego-tech 2 days ago | parent [-] | | You're both beating around the bush that is the core issue, though, and that's a lack of backup media that isn't a HDD for storing large amounts of data indefinitely, nevermind on a medium that doesn't have to be powered on every X interval to ensure it's still functional. Prosumers/enthusiasts generally have three options for large-scale data backups (18TB+), and none are as remotely affordable as the original storage medium: * A larger storage array to hold backups and/or versions as needed (~1.25x the $ cost of your primary array to account for versions) * Cloud-based storage (~$1300/yr from Backblaze B2 for 18TB; AWS Glacier Deep Freeze is far cheaper, but the Egress costs per year for testing are comparable to B2) * LTO drives ($3300 for an mLogic LTO-8 drive, plus media costs) Of those, LTO drives are (presently) the only ones capable of having a stable "shelf life" at a relatively affordable rate. Most consumers with datasets that size likely aren't reading that data more than once or twice a year to test the backup itself, and even then maybe restoring one or twice in their lifetime. LTO is perfect for this operating model, letting users create WORM tapes for the finished stuff (e.g., music and video collections), and use a meager rotation of tapes for infrequent backups (since more routinely-accessed data could be backed up to cloud providers for cheaper than the cost of an associated daily LTO backup rotation). LTO is also far more resilient to being shipped than HDDs, making it easier to keep offline copies with family or friends across the country to protect your data from large-scale disasters. It's the weird issue of making it cheaper than ever for anyone to hoard data, but more expensive than ever to back it up safely. It's a problem that's unlikely to go away anytime soon, given Quantum's monopoly on LTO technology and IBM's monopoly on drive manufacturing, making it a ripe market for a competitor. I'd still love to see someone take a crack at it though. The LTO Consortium could use a shake-up, and the market for shelf-stable tape backup could do with some competition in general to depress prices a bit. | | |
| ▲ | 0manrho 2 days ago | parent | next [-] | | I'm absolutely perplexed at how I'm beating around the bush regarding > a lack of backup media that isn't a HDD for storing large amounts of data indefinitely When I recommended LTO, which you yourself say that > LTO is perfect for this operating model Agreed. Which is why I recommended it. As did you. Because it's a solved issue. > * LTO drives ($3300 for an mLogic LTO-8 drive, plus media costs) LTO-8/9 aren't the only options. LTO5/6/7 aren't defunct/unusable/unavailable. That's like complaining that SSD's are too expensive because you're only looking at Micron 9550 or Optane P5800's and their ilk. > making it a ripe market for a competitor. You'd have to engineer your own controllers and drives and likely cartridges as well, including drivers, firmware and software, which is neither cheap nor easy which is why no one has done this. It's doable, but the initial CapEx is astronomical, and the target market outside of enterprise is small meaning the return is unlikely to make it worth it, so you'd also have to spend big on advertising to appeal to said prosumers to try to sell them on something that most people would think of as cumbersome or obsolete ("Tape?! This isn't the 80's!", sure, we know better, but does the layman? that's not an easy sell), or find someway to make inroads against IBM/HPE/Quantum in the enterprise space which is unlikely for a not-already established big name. Even in the remote chance that they can beat IBM/HPE/Quantum on $/TB on new latest gen products, they almost certainly can't do that meaningfully cheaper than buying used Quantums from a few generations ago. Would it be nice? Sure. In the meantime, price sensitive prosumers absolutely do not have to pay multiple thousands to get into tape storage. And if the data being backed up is truly that important, and we only limit it to new on shelf/current gen solutions, a one time (or once per decade or less) low 4 figures expense for a tape unit and media that is a fraction of HDD's $/TB value even at the cutting edge is not an unreasonable expense. Shit, people pay more than that for some QNAP/Synology junk with spinning disks and end up with less capacity and resilience with more headaches. If the goal is "I want to back up the data on a single HDD and don't want to spend thousands to do it" the answer is to buy another HDD and mirror/clone it. The reality is tape is still around because it already is and continues to be quite affordable (in addition to it's shelf-life/reliability), and in all likelihood (barring some breakthrough) going to outlast HDD's. | | |
| ▲ | throwaway81523 2 days ago | parent [-] | | > LTO-8/9 aren't the only options. LTO5/6/7 aren't defunct/unusable/unavailable. LTO 5 and 6 have too little capacity to be really viable these days. LTO 7 is more interesting but you're still looking at drive cost of $1000+ and media cost almost as much as HDD's per TB. |
| |
| ▲ | throwaway81523 2 days ago | parent | prev [-] | | For $1392/year you can get a Hetzner SX65 which is a fairly beefy server with 4x 22TB hard drives so that beats your Backblaze figure by about 2x, but still, it's far more than thee cost of the drives. There are also bigger models with more drives, where raid-6 overhead becomes less of an issue. https://www.hetzner.com/dedicated-rootserver/matrix-sx/ A 20TiB Hetzner StorageBox (managed Raid-6 storage with scp/Borg access) is $552 a year which is also pretty good compared to Backblaze. I have a 5TiB one and it has been solid, if a little bit slow some of the time. I think the StorageBox line is about due for a refresh since there has been a big drop in HDD prices lately, despite tariffs or whatever. Are Seagate Barracudas terrible drives, or what? They are $229 for a 20TB unit at Newegg right now. | | |
| ▲ | 0manrho 2 days ago | parent [-] | | Hetzner is a great service, but what you're pitching is in no way a solution tailored for the aforementioned usecase. We're talking long term cold storage backup medium, as in meant to last many years. SX65, Storagebox, and Backblaze are not cold storage. SX65 would be $7000 over 5 years for 80TB without redundancy. You could get an LTO-7 or even 8 drive and many times SX65's storage for less, and have literal hundreds if not thousands left over for compute or whatever else with no recurring cost. Hell you could get an autoloader all-in-one tape library with tapes to fill it for less than that. There are absolutely scenarios where SX65/Storagebox/Backblaze/Cloud-hosted-storage makes sense and is a decent value, but this isn't one of them. If you want off-site cold storage, Glacier handily beats them with money and TB's to spare. If you want always-on and available "warmer" storage, great, but that's an entirely separate discussion/usecase. | | |
| ▲ | throwaway81523 2 days ago | parent [-] | | Yes I guess for 80TB and 5 years, LTO starts looking better. For 20TB, StorageBox is still ahead. Per https://aws.amazon.com/s3/glacier/pricing/ it looks like Glacier costs $3.6 per TB per month, which is a lot more than StorageBox even not counting egress fees. Is there a cheaper class of Glacier that I missed? Another idea is simply to buy a 29TB hard drive or pair of them to keep spinning, doing occasional integrity checks. I've had terrible luck using HDD's for cold storage but by now have had a few spinning in servers for 5 years without failures. Those are hosted servers in data centers though. Environmental conditions at home might not be as good. Two HDD's idling might use $100 of electricity over 5 years. An LTO7 drive on ebay is $1000+, while new is still close to $4000. I'm dubious of used tape drives but maybe it is an ok idea. Hmm. New LTO7 tapes are around $50 (6TB capacity) so just barely ahead of HDD. I wonder why there is no "LTO as a service" where someone has an LTO drive in a data center, and for a fee will write a tape for you and ship it to you. |
|
|
|
|
