Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[TheBoy]

It could be worse, you could have an array of SD cards 

Errr... 

[zirk]

You'll be pleased to know that your OOF pages are served from SSD 

Got some 8" Floppy's kicking around if you get stuck,          dont laugh Vinyl is coming back you know   

[TheBoy]

You'll be pleased to know that your OOF pages are served from SSD 

Got some 8" Floppy's kicking around if you get stuck,          dont laugh Vinyl is coming back you know   

Not sure OOF would fit in 600kb

[Marks DTM Calib]

You'll be pleased to know that your OOF pages are served from SSD 

It could be worse, you could have an array of SD cards 

SSD is probably more reliable than HD these days as no moving parts to fail 
And obviously cheaper to run than HD's too 

Generally not.....

[Jimbob]

I had a power cut take out the boot partition of my SSD system drive a week or 2 ago! 

[omega2018]

when ssds fail (as the early ones did quite often, i had three die) they fail without any warning and fail completely.  when old style disks fail you can often get a warning if you look out for it.

[LC0112G]

I've just been looking at some data sheets for flash memory devices we use. Modern devices support > 100K erase cycles (1,000,000 typ), and > 20 years data retention. Some older devices were less than this - mainly because they were designed to replace one-time-programmable or UV eraseable EPROMS. With modern flash it's the erase cycles that usually limit SSD life.

100K erase cycles may sound a lot, but some sectors of the chip get hammered every time you write something new to the chip, and you have to erase a whole page of sectors (typically 32K or 64Kb) before you can overwrite one sector of data, so there is a lot more erasing going on than most realise. Modern driver software attempts to spread the erase/write activity over the whole chip address range, and cache write activity to reduce the number of erases required, but there are limits to what can be achieved.

I've had 2 hard disks fail this year so far. I've replaced one with a 960Gb SSD and one with a cheapo 500Gb HDD. I don't really trust SSD's and I'm not taking bets on which fails first.

All you flash gits with 8" 600Kb floppies - the ones I've got are 8" 160Kb.

[Kevin Wood]

Yep, put a flash card in a scenario where it's written to regularly and it won't last long. Does depend on the quality of the card but, generally, I'd say they are only suitable for applications where they are seldom written to.

SSDs designed for hard drive replacement often have large DRAM caches, so they are a slightly different animal to your average SD card. When they die, they do the job properly, though.

Since I started playing with raspberry Pis I have learnt to use read-only Linux systems, which gives them much better resilience.

Here's one:
:~$ uptime
 17:31:40 up 613 days,  2:02,  1 user,  load average: 0.00, 0.01, 0.05

Would never have got close to that when it was writing logs onto its' SD card.

[TheBoy]

:~$ uptime
 17:31:40 up 613 days,  2:02,  1 user,  load average: 0.00, 0.01, 0.05

Clearly a sign, Mr Wood, your Linux patching isn't, well, regular

Bit like a Solaris box I came across a work a couple of years back that had been up 20yrs .  The OS was supposedly not Y2K compliant, but clearly had no issues


I find Raspberry Pis muller SD cards.  Thanks god they are trialling netboot, bloody things.

[zirk]

I've just been looking at some data sheets for flash memory devices we use. Modern devices support > 100K erase cycles (1,000,000 typ), and > 20 years data retention. Some older devices were less than this - mainly because they were designed to replace one-time-programmable or UV eraseable EPROMS. With modern flash it's the erase cycles that usually limit SSD life.

100K erase cycles may sound a lot, but some sectors of the chip get hammered every time you write something new to the chip, and you have to erase a whole page of sectors (typically 32K or 64Kb) before you can overwrite one sector of data, so there is a lot more erasing going on than most realise. Modern driver software attempts to spread the erase/write activity over the whole chip address range, and cache write activity to reduce the number of erases required, but there are limits to what can be achieved.

I've had 2 hard disks fail this year so far. I've replaced one with a 960Gb SSD and one with a cheapo 500Gb HDD. I don't really trust SSD's and I'm not taking bets on which fails first.

All you flash gits with 8" 600Kb floppies - the ones I've got are 8" 160Kb.

160K, kind of rings a bell, I didn't even think they were that much tbh, too long ago now.

The plastic storage cases they came in were handy though, used to keep all my PC Tools in one of them when I was On Call. 

[Mister Rog]

I used to use Iomega Zip drives a hell of a lot. Brilliant at the time. I think I still have some, but can't remember last time I used them. They did used to jam now and again.

[zirk]

I used to use Iomega Zip drives a hell of a lot. Brilliant at the time. I think I still have some, but can't remember last time I used them. They did used to jam now and again.

Probably worth money now, especialy if youve got the NOC List on one of them.(Mission Impossible), dare you run it up though.

[Mister Rog]

I used to use Iomega Zip drives a hell of a lot. Brilliant at the time. I think I still have some, but can't remember last time I used them. They did used to jam now and again.

Probably worth money now, especialy if youve got the NOC List on one of them.(Mission Impossible), dare you run it up though.

Shhhhhhh . . . . . . . . . . . I know














This message will self destruct in 30 seconds