Comments on: Buying a Hard Drive this Holiday Season? These Tips Will Help

By: D Pereira

D Pereira — Sat, 02 Feb 2019 10:11:27 +0000

Before you put hard drives into your Pods to test the drives at all? Quick self-test, Long self-test, or Surface scan? Burn-in? Or just drop them in and deal with any issues that come up? I am setting up a NAS at home and got a bad batch of HGST drives that intensive surface scan weeded out – was this a good idea or am I jut putting unneeded strain on the drives?

By: hhvdblom

hhvdblom — Wed, 09 Jan 2019 12:30:36 +0000

You don’t see the PMR or SMR on the drive but its very recognizable. Most SMR drives are in Backup enclosures. The reason for this is then it seems logical you don’t put them in a NAS. Because they are in a Backup enclosure the purpose of this drive speaks for itself. By the way, after some time I dump the Backup enclosure and put the HDD in the NAS. It works the same as every drive, as long as you use it for the same purpose it was ment to be: Backup.

By: McTavish

McTavish — Sun, 16 Dec 2018 14:57:25 +0000

I don’t know of anyone else who addresses personal hard drive issues as throughly and objectively as does Backblaze, so thanks. You have solved the Cloud problem for me with Backblaze backup, but I’m still struggling with local backup.
Do you recommend any routine maintenance on spinning EHDs such as defragging. I’m a Mac user, and the utilities that I have used for defragging are pretty buggy or worse.

By: Justin

Justin — Tue, 11 Dec 2018 20:05:11 +0000

What’s a good external hard drive recommendation for a Mac? Need 2 TB drive to store photo libraries and videos. Obviously I can’t store all that stuff on the internal 256 SSD. I’m looking for decent quality. Not necessarily cheapest price. I don’t know how much a difference having a faster rpm drive might make. I would be using Lightroom or something similar to edit photos. Maybe occasionally some video editing

By: Andy Klein

Andy Klein — Tue, 11 Dec 2018 19:49:51 +0000

In reply to Stephen ♫🎹.

A couple of answers:
1) Here is the latest information we have published about the different hard drives we use.
https://www.backblaze.com/blog/2018-hard-drive-failure-rates/
2) We’ve talked previously about the parameters we use to determine if a drive has failed, but basically there are three things:
– The drive will not spin up or connect to the OS.
– The drive will not sync, or stay synced, in a RAID Array.
– The Smart Stats we use show values above our thresholds. (https://www.backblaze.com/blog/what-smart-stats-indicate-hard-drive-failures/)
The last one can be tipped off by issues with an fsck which we run regularly on all drives or through monitoring of the smart stats themselves.

By: Stephen ♫🎹

Stephen ♫🎹 — Tue, 11 Dec 2018 19:12:05 +0000

I’ve wondered the same thing as thebladrunner – how & when is a hard drive considered “failed”?
For example, is a few pending sectors sufficient (like the 38 that quickly cropped up on my 2TB 2.5″ Seagate from 2015 that I still have in my possession, but is no longer in service), or perhaps other smart errors, like the 2529 Multizone Error Rate of a WD15EADS, or 424 UDMA CRC Errors of a WD7501AALS?

I’m pretty sure a drive like this one – https://www.youtube.com/watch?v=3WnzrnL8oKA – would be considered “Failed”. How often does BackBlaze encounter a drive in a storage pod or wherever that gets like that? For example did any of the 3TB Seagates that had the high failure rates fail that way?

And speaking of Seagate, how are the newer ones, like the green label Barracuda, the Ironwolf, etc? I and my brother have been avoiding Seagate (except the one 2TB drive I got a few years ago, and the 40MB MFM drive that was in my dad’s original PC in 1989) but I wonder if I should try them again. In recent years I’ve pretty much been using HGST Deskstar NAS drives, and my older drives are mostly Western Digital Greens, with a couple Blacks and Caviars. That includes my own first drive I ever bought, which I still have (but is not in service): a WD800BB. A couple years ago it was doing the click of death, but around a year and a half ago I plugged it in again, and it revived itself, with the data intact. Interestingly, the label says WD800BB-32CAA0, but SMART in Linux and the RAID card on bootup say WD800JB-00CAA1. (For some reason, CrystalDiskInfo and GSmartControl under WIndows 10 won’t read SMART info off my Bytecc BT-PESAPA controller.)

So far the best deal I’ve gotten on a drive was two 5TB HGST Deskstar NAS units for $99 each on Black Friday 2016 at Fry’s. Most recent was a couple 10TB of the same, for I think $260 each, also at Fry’s, sometime in the past year or so I think.

By: Jim Young

Jim Young — Wed, 21 Nov 2018 05:10:09 +0000

Nice dive into the things to consider when buying a drive.

In reading it, the inveterate proofreader in me noticed a couple of things that you might want to change:

1) In the paragraph titled “CDL (component design life)” the last sentence looks to have a typo – “The is really good information, …” I assume you meant “This is …”

2) The price/GB in the table in the “What’s a Good Price” section looks to be off by a factor of 10 for the 2T – 14T drives. (It’s correct for the 1T drive.)

By: joncomputing

joncomputing — Tue, 20 Nov 2018 21:36:42 +0000

I very recently was shopping for a drive and discovered that 8TB was the sweet spot, in terms of lowest per TB cost. But only for external drives. It was actually lower than internal drives. In fact, the lowest per TB cost for internal drives was a lot smaller. But that was here in Canada.

By: DataMeister1

DataMeister1 — Tue, 20 Nov 2018 18:43:20 +0000

Why do drives 8TB and larger cost more per GB than those 6TB and less? Is that the helium tech jacking up the price, or heavier demand, or something else?