Time |
Nickname |
Message |
00:14
🔗
|
|
benjinsmi has quit IRC (Leaving) |
00:35
🔗
|
|
bithippo has joined #archiveteam-ot |
00:50
🔗
|
|
benjins has joined #archiveteam-ot |
01:48
🔗
|
|
bithippo has quit IRC (Textual IRC Client: www.textualapp.com) |
01:49
🔗
|
|
BlueMax has quit IRC (Quit: Leaving) |
02:46
🔗
|
|
BlueMax has joined #archiveteam-ot |
03:02
🔗
|
|
Panasonic has joined #archiveteam-ot |
03:32
🔗
|
|
godane has quit IRC (Quit: Leaving.) |
03:40
🔗
|
|
godane has joined #archiveteam-ot |
04:35
🔗
|
|
Mateon1 has quit IRC (Ping timeout: 255 seconds) |
04:36
🔗
|
|
Mateon1 has joined #archiveteam-ot |
04:50
🔗
|
|
odemg has quit IRC (Ping timeout: 615 seconds) |
04:54
🔗
|
t3 |
Can someone explain why tape drives are not commonly used for our archiving projects? |
04:56
🔗
|
t3 |
If it's necessary to store a lot of data, tape drives can be a useful buffer. It can be used to store a lot of data so it can be uploaded to IA at a later time. |
04:57
🔗
|
|
odemg has joined #archiveteam-ot |
05:03
🔗
|
SketchCow |
Not many of us spend the money on the kind of tape drives you mention |
05:03
🔗
|
SketchCow |
They're f'in EXPENSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSIVEEEEEEEEEEEEEEEEEEEE$$$$$$$$$$$$$$$$$$$$$$ |
05:04
🔗
|
SketchCow |
Thousands for the drive, hundreds for the tape. |
05:04
🔗
|
astrid |
might as well buy a stack of disks |
05:04
🔗
|
jrwr |
Yep |
05:05
🔗
|
|
Despatche has quit IRC (Read error: Operation timed out) |
05:06
🔗
|
jrwr |
So the nerds over at the-eye.eu are getting kicked off their host |
05:07
🔗
|
t3 |
Oh, I was thinking about how a 100 PB installation with a robotic tape drive server can be use for long term storage... |
05:08
🔗
|
t3 |
Well, there is a large upfront cost associated with that. |
05:08
🔗
|
jrwr |
tapes in most cases are good for 5-10 years, BUT they tend to rot without warning in storage sometimes |
05:08
🔗
|
jrwr |
also, damn expensive and its "offline" storage |
05:16
🔗
|
t3 |
Oh. |
05:48
🔗
|
|
dhyan_nat has joined #archiveteam-ot |
05:48
🔗
|
|
Despatche has joined #archiveteam-ot |
06:43
🔗
|
|
wp494 has quit IRC (Ping timeout: 506 seconds) |
06:43
🔗
|
|
wp494 has joined #archiveteam-ot |
06:45
🔗
|
|
killsushi has quit IRC (Quit: Leaving) |
06:58
🔗
|
|
Exairnous has quit IRC (Read error: Operation timed out) |
06:58
🔗
|
|
Exairnous has joined #archiveteam-ot |
08:47
🔗
|
|
dhyan_nat has quit IRC (Read error: Connection reset by peer) |
08:47
🔗
|
|
dhyan_nat has joined #archiveteam-ot |
09:41
🔗
|
|
MrRadar_ has joined #archiveteam-ot |
09:44
🔗
|
|
MrRadar has quit IRC (Read error: Operation timed out) |
10:20
🔗
|
|
Despatche has quit IRC (Quit: Read error: Connection reset by deer) |
10:41
🔗
|
|
BlueMax has quit IRC (Read error: Connection reset by peer) |
11:02
🔗
|
Kaz |
Oh shit |
11:21
🔗
|
|
Panasonic has quit IRC (Remote host closed the connection) |
13:22
🔗
|
|
Jopik has joined #archiveteam-ot |
13:40
🔗
|
|
Dj-Wawa has joined #archiveteam-ot |
14:48
🔗
|
|
MrRadar_ is now known as MrRadar |
15:11
🔗
|
|
VerifiedJ has quit IRC (Read error: Connection reset by peer) |
15:12
🔗
|
|
VerifiedJ has joined #archiveteam-ot |
15:33
🔗
|
|
dhyan_nat has quit IRC (Read error: Operation timed out) |
15:36
🔗
|
|
dhyan_nat has joined #archiveteam-ot |
15:45
🔗
|
|
wp494 has quit IRC (Ping timeout: 615 seconds) |
15:45
🔗
|
|
wp494 has joined #archiveteam-ot |
15:55
🔗
|
SketchCow |
But keep praying for Tape |
16:10
🔗
|
|
Dj-Wawa has quit IRC (Quit: Connection closed for inactivity) |
16:34
🔗
|
|
Stilett0- has joined #archiveteam-ot |
16:37
🔗
|
|
Stiletto has quit IRC (Read error: Operation timed out) |
19:00
🔗
|
|
VerifiedJ has quit IRC (Read error: Connection reset by peer) |
19:01
🔗
|
|
VerifiedJ has joined #archiveteam-ot |
19:10
🔗
|
schbirid |
hm, 'youtube-dl --match-title "Full Performance" https://www.youtube.com/channel/UC3I2GFN_F8WudD_2jUZbojA' wont pick up those videos, nor if i add .* around the word |
19:10
🔗
|
schbirid |
any ideas? |
19:18
🔗
|
|
Exairnous has quit IRC (Ping timeout: 252 seconds) |
19:26
🔗
|
|
Exairnous has joined #archiveteam-ot |
19:54
🔗
|
|
VerifiedJ has quit IRC (Read error: Connection reset by peer) |
19:54
🔗
|
|
VerifiedJ has joined #archiveteam-ot |
19:57
🔗
|
jut |
SketchCow: Is there a link to your Al Jazeera interview somewhere? |
20:09
🔗
|
|
Exairnous has quit IRC (Read error: Connection reset by peer) |
20:14
🔗
|
|
Exairnous has joined #archiveteam-ot |
20:21
🔗
|
|
Exairnous has quit IRC (Ping timeout: 268 seconds) |
20:26
🔗
|
SketchCow |
https://www.youtube.com/watch?v=s6jC9g6lOqs |
20:41
🔗
|
|
Despatche has joined #archiveteam-ot |
20:49
🔗
|
|
alex___ has joined #archiveteam-ot |
21:16
🔗
|
HCross |
jrwr: DP? |
21:32
🔗
|
schbirid |
jrwr: if you can edit, the audio of https://www.youtube.com/watch?v=k232xUaz2cw is super silent |
21:59
🔗
|
|
dhyan_nat has quit IRC (Read error: Operation timed out) |
22:02
🔗
|
SketchCow |
TUrn up your volume baby |
22:06
🔗
|
|
Stilett0- is now known as Stiletto |
22:10
🔗
|
schbirid |
great talk as usual |
22:10
🔗
|
schbirid |
i dont have a volume baby |
22:20
🔗
|
|
BlueMax has joined #archiveteam-ot |
22:24
🔗
|
|
schbirid has quit IRC (Remote host closed the connection) |
22:40
🔗
|
marked |
Tapes issue imho is it's cheaper per GB at first but then eventuall drives catch up and then you're stuck with an expensive tave drive than you have to upgrade expensively |
22:44
🔗
|
marked |
When doing a study on archival data for work, we isolated two desirable advantages to tape. The heads and media are separate, making recovery simpler / local damage to tape could be cut out/doesn't affect the rest of the tape. And if a SysAdmin issued rm -rf on the tape robot there would be enough time to walk over and turn it off. |
22:46
🔗
|
Kaz |
if your archival strategy includes 'turning off the tape drive when someone rm -rf's it', why even bother in the first place |
22:46
🔗
|
Kaz |
also cutting out damaged parts of tape? is this a bad joke? |
22:47
🔗
|
Kaz |
what sort of clown institution are you running |
22:52
🔗
|
marked |
I'm just repeating what was said. If you think there's something technically wrong with it, say so. Don't need to make it personal. |
22:52
🔗
|
marked |
the two claims were it was phsyically impossible for software in anyway to wipe out the archive instantly |
22:53
🔗
|
marked |
and data damage was minimized because it was localized. That tapes could be physically manipulated. |
22:54
🔗
|
marked |
however the actualy data recovery stories I heard of involved IBM sending tapes back to their secret lab |
22:55
🔗
|
marked |
if you want to call them a clown, that's your perogative, but I'll add that their data set won a nobel prize. |
23:02
🔗
|
|
alex___ has quit IRC (Quit: take care ye all. Have fun!) |
23:03
🔗
|
Kaz |
I haven't read context, is what you said a quote from somewhere, or is that what you're doing at your place |
23:04
🔗
|
JAA |
Isn't your price claim above backwards? Tape drives are really expensive, so starting with tapes is as well. But adding more capacity is cheap because the cartridges are significantly cheaper than hard drives or other media. Upgrades are of course worth keeping in mind (drives can only read the past two generations of tapes in addition to the current one, so you need to upgrade every second or third |
23:04
🔗
|
JAA |
generation), but that problem exists with other media as well. |
23:06
🔗
|
JAA |
Last time I looked into it, LTO was more expensive until you reach somewhere between 50 and 100 TB of storage, then they're significantly cheaper than HDDs. |
23:06
🔗
|
JAA |
Depends strongly on how much you pay for your drive, whether you get a new, refurbished, or used one, etc. |
23:08
🔗
|
marked |
There was no context in this thread. I was working at a university in high energy physics as a (software side) sysadmin and they were pricing the long term storage options. IBM has some common tape format that's robotisized. I presume half the safety claims came from IBM marketing but there was a story of another scientist with a corrupted tape and IBM supposedly spared no expensive that the data was |
23:08
🔗
|
marked |
recovered |
23:09
🔗
|
marked |
I think the in the end they have the IBM tape robot and some kind of hard drive array in front |
23:11
🔗
|
marked |
it's unfair to compare enterprise storag and consumer disks but there are times where consumer disks win, as Google demonstrated |
23:13
🔗
|
marked |
JAA, so my personal observation was that eventually the tape drive gets so undense that it's not worth the trouble |
23:13
🔗
|
marked |
but in a time frame that was a hassle. It also doesn't fit with our spikey data model of needing a wide pipe |
23:14
🔗
|
marked |
for a short period of time |
23:14
🔗
|
JAA |
Yeah, tape storage doesn't fit anywhere in our needs, or IA's. |
23:14
🔗
|
JAA |
It's useful to store large quantities of data which are very rarely accessed, mostly. |
23:14
🔗
|
JAA |
So offline backups etc. |
23:15
🔗
|
marked |
Facebook I just remembered was trying to get manufacturers more interested in this data access pattern |
23:20
🔗
|
marked |
hmm, what comes in Google is from 2015 using blueray |
23:20
🔗
|
marked |
https://datacenterfrontier.com/inside-facebooks-blu-ray-cold-storage-data-center/ |
23:27
🔗
|
JAA |
What's your point? |
23:30
🔗
|
marked |
Just there's a gap in the storage industry that big spenders like FB are trying to get manufacturers to fill. So hopefully there will be a new type of data storage medium that's off our radar at the moment because it doesn't exist. |
23:31
🔗
|
icedice |
Would HDDs or SSDs that are turned off survive a solar flare? Would Internet Archive survive a solar flare? |
23:32
🔗
|
icedice |
Or is everything that's not in some underground bunker be toast? |
23:32
🔗
|
icedice |
* Or would everything |
23:34
🔗
|
MrRadar2 |
I think the risk is mainly that the EMF field would induce huge eletrical currents in eletrical transmission and distribution lines. In that case anything unplugged should be fine (though good luck getting reliable electricity until the grid is rebuilt) |
23:34
🔗
|
marked |
there's no rule about OT in -ot , yeah? |
23:34
🔗
|
marked |
https://www.reddit.com/r/askscience/comments/1akw3o/could_an_emp_wipe_the_data_on_a_hard_drive/ |
23:55
🔗
|
kode54 |
are we talking a solar flare that somehow directly impacts the planet? |
23:56
🔗
|
marked |
solar flares must be inverse square over distance, I would guess? |
23:58
🔗
|
kode54 |
well, the last major solar flare happened in 2012, and it missed the earth |