I’m writing a program that wraps around dd to try and warn you if you are doing anything stupid. I have thus been giving the man page a good read. While doing this, I noticed that dd supported all the way up to Quettabytes, a unit orders of magnitude larger than all the data on the entire internet.
This has caused me to wonder what the largest storage operation you guys have done. I’ve taken a couple images of hard drives that were a single terabyte large, but I was wondering if the sysadmins among you have had to do something with e.g a giant RAID 10 array.
You must log in or register to comment.
Upgraded a NAS for the office. It was reaching capacity, so we replaced it. Transfer was maybe 30 TB. Just used rsync. That local transfer was relatively fast. What took longer was for the NAS to replicate itself with its mirror located in a DC on the other side of the country.
Yeah it’s kind of wild how fast (and stable) rsync is, especially when you grew up with the extremely temperamental Windows copying thing, which I’ve seen fuck up a 50mb transfer before.
The biggest one I’ve done in one shot with rsync was only about 1tb, but I was braced for it to take half a day and cause all sorts of trouble. But no, it just sent it across perfectly first time, way faster than I was expecting.
Never dealt with windows. rsync just makes sense. I especially like that its idempotent, so I can just run it twice or three times and it’ll be near instant on the subsequent run.
Yeah, shout out for rsync also. It’s awesome. Combine it with ssh & it feels pretty secure too.
~340GB, more than a million small files (~10KB or less each one). It took like one week to move because the files were stored in a hard drive and it was struggling to read that many files.
I’m currently backing up my /dev folder to my unlimited cloud storage. The backup of the file
/dev/randomis running since two weeks.No wonder. That file is super slow to transfer for some reason. but wait till you get to /dev/urandom. That file hat TBs to transfer at whatever pipe you can throw at it…
That’s silly. You should compress it before uploading.
Cool, so I learned something new today. Don’t run
cat /dev/randomWhy not try /dev/urandom?
😹
Ya know, if not for the other person’s comment, I might have been gullible enough to try this…
I’m guessing this is a joke, right?
/dev/random and other “files” in /dev are not really files, they are interfaces which van be used to interact with virtual or hardware devices. /dev/random spits out cryptographically secure random data. Another example is /dev/zero, which spits out only zero bytes.
Both are infinite.
Not all “files” in /dev are infinite, for example hard drives can (depending on which technology they use) be accessed under /dev/sda /dev/sdb and so on.
I’m aware of that. I was quite sure the author was joking, with the slightest bit of concern of them actually making the mistake.
It was something around 40 TB X2 . We were doing a terrain analysis of the entire Earth. Every morning for 25 days I would install two fresh drives in the cluster doing the data crunching and migrate the filled drives to our file server rack.
The drives were about 80% full and our primary server was mirrored to two other 50 drive servers. At the end of the month the two servers were then shipped to customer locations.
A few years back I worked at a home. They organised the whole data structure but needed to move to another Providor. I and my colleagues moved roughly just about 15.4 TB. I don’t know how long it took because honestly we didn’t have much to do when the data was moving so we just used the downtime for some nerd time. Nerd time in the sense that we just started gaming and doing a mini LAN party with our Raspberry and banana pi’s.
Surprisingly the data contained information of lots of long dead people which is quiet scary because it wasn’t being deleted.
No idea about which specific type of business it is, but keeping that history long term can have some benefits, especially to outside people. Some government agencies require companies to keep records for a certain number of years. It could also help out in legal investigations many years in the future and show any auditors you keep good records. From a historical perspective, it can be matched to census, birth, and death certificates. A lot of generational history gets lost.
Companies also just hoard data. Never know what will be useful later. shrug
80GB, it was 8 hours of (supposedly) 4k content in the MP4 format. https://www.youtube.com/watch?v=VF5JWdaJlvc Here’s the link (hoping for the piped bot to appear).
I work in cinema content so hysterical laughter
Interesting! Could you give some numbers? And what do you use to move the files? If you can disclose obvs
A small dcp is around 500gb. But that’s like basic film shizz, 2d, 5.1 audio. For comparison, the 3D deadpool 2 teaser was 10gb.
Aspera’s commonly used for transmission due to the way it multiplexes. It’s the same protocolling behind Netflix and other streamers, although we don’t have to worry about preloading chunks.
My laughter is mostly because we’re transmitting to a couple thousand clients at once, so even with a small dcp thats around a PB dropped without blinking
Ahhh thanks for the reply! Makes sense! We also use Aspera here at work (videogames) but dont move that ammount, not even close.
I used to work in the same industry. We transferred several PBs from West US to Australia using Aspera via thick AWS pipes. Awesome software.
Hahahah did you enjoy Australian Internet? It’s wonderfully archaic
(MPS, Delux, Gofilex or Qubewire?)
In the early 2000s I worked on an animated film. The studio was in the southern part of Orange County CA, and the final color grading / print (still not totally digital then) was done in LA. It was faster to courier a box of hard drives than to transfer electronically. We had to do it a bunch of times because of various notes/changes/fuck ups. Then the results got courier’d back because the director couldn’t be bothered to travel for the fucking million dollars he was making.
Fucking hell the raws woulda been gigantic
You legally have to tell us if that movie was Shrek.
Hah, nope. Shrek was made in Glendale, so they probably had everything on site or right next door.
Oh yeah I worked in animation for a bit too. Those 4K master files are no joke lol
Eh, what’s a dcp?
Digital Cinema Package. Films come out in a buncha files that rather resemble a dvd rip. You got your video files (still called reels!) and your audio files, maybe some subtitle files and other bits and pieces and your assetmap (list of files) all in a big fat folder collectively called a DCP
Digital Cinema Package; basically the movie file you’re watching when you’re in a movie theater.
That article was a weird mix of insider info and wild inaccuracies
Oh sorry! Here ya go!
When I was moving from a Windows NAS (God, fuck windows and its permissions management) on an old laptop to a Linux NAS I had to copy about 10TB from some drives to some other drives so I could re-format the drives as a Linux friendly format, then copy the data back to the original drives.
I was also doing all of this via terminal, so I had to learn how to copy in the background, then write a script to check and display the progress every few seconds. I’m shocked I didn’t loose any data to be completely honest. Doing shit like that makes me marvel at modern GUIs.
Took about 3 days in copying files alone. When combined with all the other NAS setup stuff, ended up taking me about a week just in waiting for stuff to happen.
I cannot reiterate enough how fucking difficult it was to set up the Windows NAS vs the Ubuntu Server NAS. I had constant issues with permissions on the Windows NAS. I’ve had about 1 issue in 4 months on the Linux NAS, and it was much more easily solved.
The reason the laptop wasn’t a Linux NAS is due to my existing Plex server instance. It’s always been on Windows and I haven’t yet had a chance to try to migrate it to Linux. Some day I’ll get around to it, but if it ain’t broke… Now the laptop is just a dedicated Plex server and serves files from the NAS instead of local. It has much better hardware than my NAS, otherwise the NAS would be the Plex server.
so I had to learn how to copy in the background, then write a script to check and display the progress every few seconds
I hope you learned about terminal multiplexers in the meantime… They make your life much easier in cases like this.
30 years with Linux and I know I still haven’t. Maybe this year? :-D
I don’t remember how many files, but typically these geophysical recordings clock in at 10-30 GB. What I do remember, though, was the total transfer size: 4TB. It was kind of like a bunch of .segd, and they were stored in this server cluster that was mounted in a shipping container for easy transport and lifting onboard survey ships. Some geophysics processors needed it on the other side of the world. There were nobody physically heading in the same direction as the transfer, so we figured it would just be easier to rsync it over 4G. It took a little over a week to transfer.
Normally when we have transfers of a substantial size going far, we ship it on LTO. For short distance transfers we usually run a fiber, and I have no idea how big the largest transfer job has been that way. Must be in the hundreds of TB. The entire cluster is 1.2PB, bit I can’t recall ever having to transfer everything in one go, as the receiving end usually has a lot less space.
4G?! That strikes fear into my heart!
The alternative was 5mbit/s VSAT. 4G was a luxury at that time.
At the rates I’m paying for 4G data, there are very few places in the world where it wouldn’t be cheaper for me to get on a plane and sneakernet that much data
Why would dd have a limit on the amount of data it can copy, afaik dd doesn’t check not does anything fancy, if it can copy one bit it can copy infinite.
Even if it did any sort of validation, if it can do anything larger than RAM it needs to be able to do it in chunks.
It’s less about dd’s limits and more laughs the fact that it supports units that might take decades or more for us to read a unit that size.
Not looking at the man page, but I expect you can limit it if you want and the parser for the parameter knows about these names. If it were me it’d be one parser for byte size values and it’d work for chunk size and limit and sync interval and whatever else dd does.
Also probably limited by the size of the number tracking. I think dd reports the number of bytes copied at the end even in unlimited mode.
Well they do nickname it disk destroyer, so if it was unlimited and someone messed it up, it could delete the entire simulation that we live in. So its for our own good really.
No, it can’t copy infinite bits, because it has to store the current address somewhere. If they implement unbounded integers for this, they are still limited by your RAM, as that number can’t infinitely grow without infinite memory.
Multiple TB when setting up a new server to mirror an existing one. (Did an initial copy with both together in the same room, before moving the clone to a physically separate location. Doing that initial copy would saturate the network connection for a week or more otherwise)
I once moved ~5TB of research data over the internet. It took days and unfortunately it also turned out that the data was junk :/
While I haven’t personally had to move a data center I imagine that would be a pretty big transfer. Probably not dd though.
I can’t imagine how nerve-wracking it would be to run dd on something like that lol. I still don’t trust myself to copy a USB stick with my unimportant bullshit on it with dd, let alone a server with anything important on it!
I’ve imaged an entire 128GB SSD to my NAS…
In the middle of something 200tb for my Plex server going from a 12 bay system to a 36 LFF system. But I’ve also literally driven servers across the desert because it was faster than trying to move data from one datacenter to another.
That’s some RFC 2549 logic, right there.
Just thinking about how much data you could transfer using this. MicroSD cards makes it a decent amount. Latency would be horrible, but throughput could be pretty good I think.
Amazon Snowball will send you a semi truck.
Packet loss would be quite costly though
Which desert? I’ve lived in the desert my entire life.
From LA to Vegas. Took the servers down end of business one night, drove it all night, installed it and got it back online before start of business the next day.
As an ex-Vegas resident, I have to ask: why were you moving stuff to Vegas?
It’s got a hell of a datacenter.
