![]() These indexes scale with the number of blobs (data chunks) so this approximately scales with the total size of your backup set, but is also affected by how many individual files are in the repository. ![]() This is approximately the number PUT requests for pack files. So for packs, take your total data set (absent deduplication – we’ll assume this is a first-time backup with minimal duplication) and divide it by 8MB. Restic targets pack files to be around 8MB, and all of your data will go in packs. A lock file is created and periodically updated (I think once every 5 minutes) so the backup duration divided by 5 minutes = the number of lock puts.LTO 7 speed is 300MB/s (raw) compared to closer to 60MB/s for a standard hard drive (5400 will be much slower, 7200 about this, 10K a little faster.) But an LTO 7 will crush even a 15K hard drive in write performance here.Four things get written when you run a backup: If you do anything else, it'll slow down a lot. And that's only if you write sequentially. The backup window has to be huge just to write to the drive. ![]() Filling 6TB will likely take most of a day, if not an entire day. If capacity isn't your only issue, performance likely is. Now we are really starting to see savings. So in real world terms, the LTO is $4/TB compared to $18.33/TB for the HD. So realistically, no one talks about raw capacity on LTO because the compression is part of the tape spec and you expect to get 15TB on a drive, compared to actually getting 6TB on the hard drive. And pretty much no compression meets the hardware compression that LTO does, it's about the best that there is. You want to compress what is going onto it, you have to buy and deal with that separately. That HD is raw and that's all that you get. And that's enterprise LTO vs "cheapest consumer item on Amazon" HDD. So in raw capacity, the LTO is roughly half the cost of a HD. What do you think about their Assumptions page? I think most companies will make do with what they have and will not experience these assumed costs. My biggest issues are with Annual personnel cost and Network Cost. Network Cost (monthly) $400 Incremental cost to upgrade network speeds (symmetric 500 Mbps minimum) Wage growthĒ%Ěnnual wage increase for LTO personĬost of Storage $0.005Ĝost of storage per month per GB (Backblaze)ĭownload percentageđ0% Percentage of data downloaded each year, based on the total amount of dataĬost of Download $0.01Ĝost to download 1 GB of data Here's a piece of their "assumptions" page:Īnnual personnel cost $104,000Ěnnual salary + benefits, FTE who manages LTO system(s) - $50/hour Their calculator doesn t apply in the IT world I have worked in for more than 30 years. If their goal was to compare "Perfect" scenarios with "Perfect" business models with "Perfect" cost analysis, then I'll give them credit. I think their calculator is highly flawed for use in the real world of existing customer who already have some sort of backup in place. BUT the fact of the matter is, most companies will not.įor example, if a company is going to ask an EXISTING employee to manage those backups, they are not going to calculate the cost of labor for the backup in their decision, they are going to add that task to the existing workload of that employee and not change the compensation. I know people will say you have to include those costs, or you are not seeing the real price yes I know that. ![]() ![]() Or at least would never be included in many business decisions. They make some assumptions that just don't apply to many businesses. ![]()
0 Comments
Leave a Reply. |