Known-good drive manufacturers, known-trusty PC review sources

Quis custodiet ipsos custodes?

Do you have a better suggestion for economical mass storage?

jsw

When I was a test engineer the standard advice was to run or "burn in" hi-rel devices for a week to catch infant mortality. Shippable product ran at room temperature, small sacrificial lot samples were temperature cycled between the industrial or military extremes. In practice most product received only a short functional test that didn't slow down the production line. The S.M.A.R.T data in a new blank drive doesn't show evidence of much testing.

I run a read and transfer rate tests a few times, fill the drive and test it again, then open the program that uses the data and check a few files for faults.

With a laptop executing the tests the electricity cost is around $0.10 a day. My fast desktops draw nearly ten times the power. jsw

My home data management scheme:

Primary storage - LG NAS box with two 1TB disks in a mirrored configuration. Fault tolerant and supporting storage for all important files accessed by the 7 or 8 PCs here. Backups of the PCs systems disks are also pushed here periodically. Everything is on UPSes of course.

Offline onsite storage - 1TB Seagate USB3 portable drive, encrypted with TruCrypt and kept in a firesafe. This is a full copy of the LG NAS.

Offline offsite storage - 1TB Seagate USB3 portable drive, encrypted with TruCrypt and kept in a safe deposit box at a bank a few miles away. This is a full copy of the LG NAS.

Periodically the offline onsite drive is retrieved from the safe, updated with a fresh encrypted copy of the NAS and then brought to the bank to swap out with the one in the safe deposit box. The one from the box is brought back, updated and put in the safe.

Simple, inexpensive (~$600 in gear), reliable and easy to manage.

My current idea of good practice is to maintain a clean OS master on a second plug-in bootable hard drive that I update and then clone to the primary drive. Occasionally I make a bootable DVD DL backup of the master that I could use to restore a corrupted or infected primary drive without exposing the master drive to infection.

Although it's out of date it puts a known-good system on the primary so I can then restore the latest backup from a hard drive. Restoring the XP or 7 OS partition over USB2 takes about 15-20 minutes.

On laptops and those desktops that won't hold two physical drives I have the OS and all data on separate partitions. I bought CD-bay hard drive caddys for my Dell laptops and swappable drive cartridge holders for my multi-CD-bay desktops.

A USB external drive would also work but they are slower and not useful for transfer rate testing, while the CD bay drive runs at full UDMA5 speed or its own best, whichever is lower. The HDTune Benchmark scan I posted shows how transfer rate can quickly reveal a weak drive that passes Check Disk. The bootable second drive can test the primary without showing Windows overhead glitches.

jsw

One of my Seagate 1TB USB3 portables has a problem with intermittent connection. So far wiggling the adapter has made it work fine until the next time I plug it in. On the 2T desktops the USB2 adapter base seems more reliable than the USB3 one, and they can be swapped among drives. jsw

None at all. For a very good reason too.

Well that autoloader selects from 16 tapes @ 1.5 TB per tape. You have

100s of tapes but at what size per? What total TB?

Oh, and your data de-dup has long been known as incremental backup. That has interesting implications for a bare metal restore.

?-)

Using "snapshots" the backup is as simple as with a full backup.

Way, way past TB. Dunno, latest greatest Storagetek libraries that are somewhere around 8' wide by 16' long. I don't manage them so I don't know the details other than they probably cost more than my annual salary.

Data de-dup is light years ahead of incremental backups. Incremental backups work at the file level, data de-dup works on the contents.

Snapshots just let you backup stuff like a large database without having it unavailable for more than a second or two while you spin out a mirror set member to use for offline backup. Nothing to do with data de-dup.

nah de-dup is different - moderrn datacentres often end up with several copies of the same file on the (live disk),

incremental backups needn't be hard to restore. disks are random access...

we do overlapped backups such that todays backup shares the files in yesterdays when they are identical.

--
?? 100% natural 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

de-dup.

have

That

OK so it adds file content level forward deltas, ala RCS et seq. It may save a lot of backup media space, but at an ugly cost at restore time, especially a bare metal restore.

Sitll not all that impressed. There were scripts doing that in Unix and similar OSs two decades ago.

?-)

have

That

I dare you to do a full restore from bare metal some weekend. Nothing critical, it just has to have been running for over a year.

?-)

Took me 56 hours the last time - don't EVER want to go through that again

what does uptime have to do with it?

the way I do it the content of the most recent incremental backup is a full backup.

--
?? 100% natural 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

I've done it a few times with my (old) Lenovo laptop, though certainly not by choice. The first time it was scary but also nothing to lose. I had perfect results each time.

I have two or more copies of about 3 TB of incompletely organized backups, some incremental. Replacing a boot drive takes perhaps half an hour because it doesn't hold any data. Replacing a failing backup drive is a separate off-line operation on a laptop with a dual USB3 Expresscard, set up wherever I can clear temporary safe space for the laptop and backup drives, often on the washing machine. It seems to read from simultaneously connected USB2 drives faster than it writes to them, maybe because a file read doesn't update the directory. Friday I had 8 TB plugged into the laptop, to reorganize and remove extra outdated copies. jsw

I'll bet it wasn't running 4 SQL databases and exchange server, with

28 users. One of the SQL databases is 140GB. 5 virtual machines in total - all the SQL on one, Exchange on another -

have

That

several

full backup.

No it isn't. That is what i am trying to tell you.

Read snipped-for-privacy@snyer.on.ca's post. 56 hours is NOT a joke.

?-)