• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

WD "Red" drives? At Newegg.

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I've had many WD Green drives in my server over the past few years. The 1TB drives that I retired and sold two years ago had well over 1.5 million load/unload cycles according to SMART. I think most people who had WD Green drives fail looked at the load/unload cycle count and realized it was significantly higher than a "traditional" hard drive.

Correlation != Causation
 
I have a WDC WD7500BPKT-00PK4T0 Scorpio Black 2.5" laptop HDD that is 16 months old and as of today:
Code: 
193 Load_Cycle_Count 721120

The threshold is 000, and the value is 001, although the manual specs 600,000 cycles for this this drive (as a minimum so the value may never go below 1).

Anyway, it's not just the green drives that exhibit this behavior.
 
SPCR reviews 1TB and 3TB WD Red

They recommend it for the following reasons:

Both the 3TB and 1TB version are the most energy efficient and quiet models we've tested in their respective capacities. Seek noise is almost nonexistent and vibration levels are excellent.
...The 3TB variant however, is much faster, outclassing older 7200 RPM models like the 2TB WD Caviar Black and Seagate Barracuda XT in our real world application tests. If you want it all — performance, a minimal noise footprint, and a respectable amount of capacity, the WD Red 3TB is about as good as it gets, at least for a single, one-drive fits all solution. It doesn't compare to a solid-state drive or the latest 1TB VelociRaptor but it's speedy enough that you'd be hard pressed to notice the difference compared to a standard 7,200 RPM model.
Click to expand...
 
They're a bit difficult to get at the moment. Newegg presently has a 2 quantity limit. Looks like this is WDC's first 7200 RPM 3TB disk under its own label. Hitachi has a 3TB Deskstar also at 7200RPM but a 3 year warranty. To get a 5 year warranty, that's the 3TB Ultrastar which comes in SAS and SATA. That is an enterprise disk though. So the WDC Red is an interesting middle ground.
 
Oops I'm wrong. The WDC Red is a 3 year warranty. So it's worth considering the Hitachi Desktar which is also at a 3 year warranty. But I'm not seeing Deskstar sustained performance spec, only Ultrastar.

Sustained spec:

3TB Ultrastar SAS 157MB/s
3TB Ultrastar SATA 166MB/s
3TB WDC Red SATA 145MB/s
 
WDC Red is also an AF disk (512e). The Deskstar uses 512 physical sectors. Some performance info here. Looks like this Deskstar is up there with the Velociraptor performance wise. Same 3 year warranty, and 24/7 usage.

There is also a 4TB 7200 RPM Deskstar, which is an AF disk, the 7K4000.

The Red has an advantage on power consumption, 4.1W idle vs Deskstar's 6.9.
 
ViRGE said:
So good news.

It looks like jumping pins 7-8 on a Red drive will apply a sector offset, just like it does on WD's Green drives.

I installed the 2TB Red this evening, and before adding it to the drive pool ran some sequential write tests on it with both CrystalDiskMark and IOMeter. Both are showing 90MB+.

On a side note, it's also possible to disable TLER using the WDTLER utility.
Click to expand...

This is great news. I have a WHS v1 that needs a 2TB drive. I have been using Advanced Format drives and just jumpering them. What is TLER? I have tried running WD utilities before in WHS, against Advanced Drives and have never been successful. I forgot what I was trying to accomplish but the utility never was able to interface with the drives correctly and make the change or set the setting.
 
The whole idea of these drives for a NAS is if you're using it in the recommended products, those products will configure the drive if it needs either of these features.

TLER will cause the drive to report a CRC error more quickly when encountering a sector read error, rather than continuing to make attempts. In RAID 0, or single disk use, this is not a great idea. It's better for the drive to error correct because there is nothing above the disk level that can correct for it. Whereas with RAID 1, 4, 5, 6 or RAIDZ there is at least the possibility of correcting for a failed sector. The implementation is what will determine whether it actually does or not.

It's kinda hard to test these sorts of things, because there isn't a way to get the drive to, on command, issue a CRC error and see what the handling is by layers above (RAID and file system). i.e. would software RAID (including mdraid) either 1, 4, 5, or 6, automatically use the mirrored copy of a CRC'd sector, or rebuild it from parity? And if it rebuilds from parity, does it then attempt to write out that chunk forcing the drive to write those sectors, which would cause any persistently bad sectors to be automatically removed from use by drive firmware?

I don't know. But in theory that's how it would work. It would be nice to know what happens (and no doubt it depends on the RAID implementation).
 
murphyc said:
It's kinda hard to test these sorts of things, because there isn't a way to get the drive to, on command, issue a CRC error and see what the handling is by layers above (RAID and file system).
Click to expand...

On Linux, the md driver has a FAULTY module that allows such testing simulation. And in fact, the md driver does correct for read errors, and write over those bad sectors and re-reads them. If the write or re-read fails, the device is failed if it's RAID 1456.
 
EJ257 said:
WD copying the Johnnie Walker labeling scheme.

Johnnie Walker
Green Label
Blue Label
Black Label
Red Label
Gold Label

WD
Caviar Green
Caviar Blue
Caviar Black
NAS Red

WD just need to slap a gold sticker on the RE4 and the collection is complete 😀
Click to expand...

Brilliant!

however in JW, Blue is the most expensive not BLACK!! 😀
 
so the main reason not to use a red over a green in single drive storage is the way it handles error correction right? is it possible to use western digitals tools to change the error correcting to that of a green drive?

just wondering since i cant get either drive for the same price.

also in raid1 is tler beneficial?
 
vibratingdonkey said:
so the main reason not to use a red over a green in single drive storage is the way it handles error correction right?
Click to expand...

Presumably the Red has an adjustable ERC, as enterprise disks do, so you could ratchet it up so that it's in the realm of a Green. Otherwise, you could not want to use a Red in a non-RAID situation, with default ERC, because it will quickly just give up and start throwing sector read errors, which in a single disk situation will cause all kinds of mayhem.

is it possible to use western digitals tools to change the error correcting to that of a green drive?
Click to expand...

I'm pretty sure the Red allows TLER adjustment (check the smartmontools man pages for scterc to find out).

You may very well want to do raise the TLER value (or disable it entirely) if one drive dies in RAID 1 and 5, or once a second drive dies in RAID 6. After the critical number of disks fail, any sector error translates into data loss, so more aggressive ERC at this point is more desirable.

People with mission critical situations, truly mission critical, will have a cluster file system with replicated storage bricks. So such a storage brick that starts going wonky, with lots of delays retrieving data, is going to be demoted out of the cluster until its array(s) are non-degraded.

also in raid1 is tler beneficial?
Click to expand...

It depends. If you're using a RAID controller that will time out and eject a disk out of an array when a disk goes into deep recovery, then yes a WDC disk with TLER is beneficial in that the delay will be less, the controller will correct for the error instead of dropping the disk. However, once a disk is experiencing 7+ second recoveries for sectors, it's in trouble. Why isn't this disk being monitored by smartd? Why isn't this disk being periodically scrubbed? There are maintenance tasks that will help avoid this ridiculous TLER "problem".

And by the way it's ERC. TLER is a WDC marketing term for their version of it.
 
BFG10K said:
It's not 7200RPM, it's 5400RPM (approx).
Click to expand...

How do you know this?

I just asked myself this question, and thought I'd read it before writing that, but the specs clearly say IntelliPower for these Reds, and a seek time isn't published either.
 
At this point, baring additional information, I'd sooner use the Hitachi Deskstar which reports indicate still have an adjustable SCTERC setting; or the Hitachi Ultrastar if you want an enterprise SATA or SAS disk.
 
Zxian said:
The sound they produce indicates the rotational speed of the spindles. Greens and Reds are 5400RPM.
Click to expand...

Very good catch. I must have gotten confused over the sustained transfer benchmarking indicating it was doing as well as 7200 RPM drives.
 
murphyc said:
Very good catch. I must have gotten confused over the sustained transfer benchmarking indicating it was doing as well as 7200 RPM drives.
Click to expand...
If nothing else, look at WD's lineup. It wouldn't make any sense for the Reds to be 7200 RPM, as WD already has such a drive: the RE4 series. The Reds are effectively 5400 RPM versions of RE4, or an RE4ized version of the Green depending on how you want to look at it.
 
murphyc said:
How do you know this?
Click to expand...
  • No drive that runs at 7200RPM or faster has its speed masked with marketing; they outright say it.
  • At 1TB/platter, sustained transfer rates are too low for it to be running at 7200RPM. 7200RPM drives get around 150MB/sec average with similar platters.
I just asked myself this question, and thought I'd read it before writing that, but the specs clearly say IntelliPower for these Reds, and a seek time isn't published either.
Click to expand...
You can't really go off seek times as some 7200RPM drives have high latency too (e.g. Hitachi 7K1000.D).

Very good catch. I must have gotten confused over the sustained transfer benchmarking indicating it was doing as well as 7200 RPM drives.
Click to expand...
See above. Sustained transfers look good because the 7200RPM drives they compared it to are using last generation platters, which have low areal density in comparison. If they put a 7K1000.D, ST1000DM003 or WD10EZEX in there, the results would look a bit different.
 
Would a WD Red drive make a good second system "storage" drive with the primary drive being an SSD for OS, applications, etc.? Basically, put a Red drive and an SSD in a standard PC enclosure instead of a NAS.

I like the low noise and low power consumption, but is this drive too tailored for a standard role?
 
jeffrey said:
Would a WD Red drive make a good second system "storage" drive with the primary drive being an SSD for OS, applications, etc.? Basically, put a Red drive and an SSD in a standard PC enclosure instead of a NAS.

I like the low noise and low power consumption, but is this drive too tailored for a standard role?
Click to expand...
No. A Red and its higher associated cost only make sense for a NAS. For what you want to do, a Green would be just fine.
 
ViRGE said:
No. A Red and its higher associated cost only make sense for a NAS. For what you want to do, a Green would be just fine.
Click to expand...

Thanks ViRGE!
Yeah, that's the dilema, the 2TB Green is just under $100, the Red is around $130, and Amazon has the Black right now for $144. I was wondering if the 1TB platters on the Red series are a worthy upgrade over the Green drives to help bridge the gap in performace between Green and Black. I know the Black is fastest, but the FAEX is showing it's age in platter density compared to the Red drives. If the Red's 1TB platters could improve performace over Green with lower noise and power than Black, then there might be $30 worth of value there.
 
Would using Red drives make sense for a 5- or 6-drive RAID array where each drive is plugged directly into the PC's motherboard? My objective is capacity and longevity (i.e., I don't want drives failing after a year or two), not speed. I still haven't decided between on-board hardware RAID or Windows soft RAID. My current setup uses 5x green-type drives (from mixed vendors) in an external RAID enclosure, and I'm very worried about the drives starting to fail. Plus I'm looking to get rid of the enclosure itself.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top