FORUMS: list search recent posts

G Speed Q ESATA Write Speeds in Raid5

COW Forums : RAID Set-Up

<< PREVIOUS   •   VIEW ALL   •   PRINT   •   NEXT >>
Kevin Colber
G Speed Q ESATA Write Speeds in Raid5
on Aug 30, 2013 at 9:55:40 pm

Hey gang- can anyone give me any info on what their write speeds on a G-Speed Q in Raid5 are?

I'm only seeing like 95 MB/s Write. Read is good at 245MB/S.


We are running these (tested on 2 different units) ESATA into a Fasta-6GU3 in a Sonnet Echo Express Pro TBolt box.


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Aug 31, 2013 at 8:04:17 pm

What model drives?


Return to posts index

Rainer Wirth
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 2:03:31 pm

Speed with esata varies, depending on how full your drives are. If your drives are filled up let's say 30 % you should get around 180-200MB/s if the drives are filled up more than 70% speed is going down to less than 100 MB/s.

cheers

Rainer

factstory
Rainer Wirth
phone_0049-177-2156086
Mac pro 8core
Adobe,FCP,Avid
several raid systems


Return to posts index


Kevin Colber
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 2:13:59 pm

This thing is nearly empty and we are seeing 95MB/s write. Read is great at 250MB/s based off of the AJA and Blackmagic tests.


Return to posts index

Kevin Colber
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 2:18:55 pm

This thing is nearly empty and we are seeing 90MB/s write. Read is great at 250MB/s.

Drives are HGST 4TB 7200 SATA 6.


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 7:25:38 pm

If it's the HUS724040ALE640 then I think there is more than one thing going on that's not right, because those drives should do ~180MB each, so striped reads should draw a lot more than 250MB/s depending on the specifics of the configuration. The write speed could either be low for the same reason the read is, which is an chunk size that's not appropriate for the workload; or it could be excessive RWM due to misalignment, as these drives are 512e AF disks.

So more information is needed. Exact model number. What's the raid card. What's the chunk and/or stripe size. How many disks. What is the work load (how big are the files being copied when the 95/250 benchmark is determined.)


Return to posts index


Kevin Colber
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 8:30:19 pm

The drives are:

HUS724040ALE640, 4 disks.


I've done the write test at 2gb, 4gb, and 8gb.


Can you tell me how to check the other options?


Return to posts index

Kevin Colber
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 9:01:27 pm

Some more tests:

FW800:

2.0GB, 65MB/s Write, 81MB/s Read
8.0GB, 60MB/s Write, 82mb/s Read


ESATA:

128MB 59/234
512MB 98/245
1GB 98/249
2GB 99/238
4GB 97/250


For reference this is a G-Tech G Speed Q 16TB Raid 5


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 9:30:17 pm

For chunk size, the goal is to maximize full stripe writes. Full stripe is chunk size * number of data disks. To find chunk size you need to look for a chunk or stripe size value in your raid software. And you need to know something about your typical write sizes, to figure out chunk size. A too high chunk size gets hit with RWM penalty at the raid chunk level. Too small chunk size means the disks become excessively bound in IO which also hurts performance.

Another factor is the HFSJ/HFSX journal. It's 8MB per 100GB of file system up to 512MB. So you have a 512MB journal which is quite large. It might be worth disabling the journal and redoing your tests to see if there's any change. Ultimately you want the journal on to avoid either long fsck times in the event of a crash, or not doing fsck at all in which case it's a matter of not much time before the file system trashes itself. It can be temporarily disabled through the diskutil command. If performance picsk up dramatically then you probably have moderate to heavy metadata writes, and to fix this longer term either reduce the chunk size, or relocate the journal to an SSD, which is supported but not in the GUI.

Another source of RWM performance hit is when the raid chunks aren't aligned to a drive's physical sectors. This ought to not happen these days. But it depends on if the disks are partitioned or not and what partitioned them and if it does so correctly. This is not the same thing as partitioning the resulting array that appears in Disk Utility. The raid manufacturer should be able to tell you this.


Return to posts index


Alex Gerulaitis
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 3, 2013 at 9:29:46 pm

[Kevin Colber] "Hey gang- can anyone give me any info on what their write speeds on a G-Speed Q in Raid5 are?"

Haven't tested it myself. Similar 4-bay multi-interface RAID5 boxes from CineRAID and ProAvio - read speeds similar to yours; write speeds in the 150MB/s range.

G-Tech used to post benchmarks on their product pages - don't see any for the "Q". Bummer.

The only speed test I saw for the "Q" has numbers similar to yours (on 3G eSATA): 174/104. The review is from two years ago though, with a different Q model - the one that didn't have USB 3.0 yet. (The funny part is that even their current datasheet still has USB 2.0 on it.)

It's unlikely anyone here will be able help you no matter the alphabet soups they throw at the problem: the only way to get real answers is to ask fellow Q users, what numbers they are seeing, and you may have better chances on G-Tech forums.

My hunch is that what you're seeing is normal, the effect of either a slow RAID5 brain, or having write caching off on the drives or on the controller. (The latter is the right way to do it with RAID5 controllers w/o battery backup.) I could be wrong of course. The Q however was never intended a speed demon despite the "speed" moniker: they're primarily backup and archival devices.


Return to posts index

Rainer Wirth
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 4, 2013 at 2:36:38 pm

Alex is right,

these are quite normal parameters for esata. For editing the read parameter is much more important than the write speed. If the array is filled up, the speed goes down significantly, that's why we use esata for storage only.

cheers

Rainer

factstory
Rainer Wirth
phone_0049-177-2156086
Mac pro 8core
Adobe,FCP,Avid
several raid systems


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 4, 2013 at 6:39:31 pm

The issue isn't eSATA, which is the exact same thing as SATA except for the physical connector. Protocol and bandwidth are the same. Also eSATA neither correlates nor causes slow down when the array starts to fill up. Array slow down is a function of file system fragmentation, which also leads to heavier raid layer RWM penalties.

The bandwidth of the eSATA connection in this case is 600MB/s after accounting for 8b/10b encoding overhead. The drives can do large sequential writes at ~135MS/s per drive. The stripe width is 3 in this case, which makes for 405MB/s full stripe writes: assuming the raid hardware in these units can do that, and the layout is optimized for the workload, any of which may not be true. But the performance hit isn't due to eSATA or the drives, and probably isn't due to the Fasta card or the PCIe-TB enclosure or the bandwidth limits of Thunderbolt compared to a legit 4-lane PCIe slot.

I think it's either misaligned chunks to 4K physical sectors of the drive causing drive firmware RWM, and/or the lack of full stripe writes causing excessive RWM of chunks at the raid layer. And the raid hardware isn't powerful enough to optimize its way out of that given the workload.

I just looked at the user manual for this box and it doesn't show a configuration utility option for setting chunk size or layout options. Rather than powering down the unit, pulling a drive, and putting it on a separate SATA bus, and looking at sector data to figure these things out, it's easier to just ask G-Technology support what the chunk size is, and confirm/deny that their firmware supports proper alignment of chunks on 4K physical sector drives.

I'd also question the testing method that comes up with 95MB/s writes, which may not reflect the intended actual real world usage. If the workload between the test and the real world aren't the same, the test is useless. It may indicate better or worse than real world performance.


Return to posts index


Rainer Wirth
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 4, 2013 at 9:56:42 pm

so Chris,

if you are so fond of esata and it is terribly speedy, why do we end up with thousands of bucks spending in a FC array?
Our experience has showed us, that esata is not the speedyist solution on the market and not that reliable when it come sto speed and a 70% filled up array.
I think, the parameters showing with AJA are correct and underline our experiences.

cheers

Rainer

factstory
Rainer Wirth
phone_0049-177-2156086
Mac pro 8core
Adobe,FCP,Avid
several raid systems


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 4, 2013 at 10:43:20 pm

I did not state a like or dislike of SATA or eSATA. I stated that its available bandwidth, in this thread's context, is a non-factor. Given the read is 225MB/s which is below the SATA Rev 2.0 3Gbps bandwidth, it makes no sense to suggest FC alone would solve this problem. The bandwidth between array and host is not the bottleneck, it's something else.

FC to SATA is apples to oranges. Totally different available bandwidths by year, different protocols SCSI vs ATA and hence different command queuing and ECC, and FC implies SAS drives so different mechanisms possibly faster rotational rates. It's like, nothing is the same between them. But it wouldn't matter if the G Speed Q had an FC port capable of 8GFC, the host to array bandwidth is not being saturated now as it is.

I think your experience is that you've used unreliable SATA cables, drives, cards, raids, and as for the 70% full business again that has nothing to do with the connector being used. The connector knows nothing about how full a hard drive is. A file system does, and the raid RWM processing will be affected by this as well.


Return to posts index

Rainer Wirth
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 4, 2013 at 11:12:18 pm

But Chris,

how is it, that we experience huge speed loss with esata connections in comparison with let's say SAS connections. In theory you could be right, but daily practise tells us another story. We've tested this under various conditions - it's all the same - esata doesn't provide us as a production house with enough speed we need for online editing. We therefore have decided that SAS and FC are the only solutions for us. In the future Thunderbolt might be an answer, but even thunderbolt doesn't reach the speed of a 4x8GB/s FC Raid. In this thread someone is coping with the actual speed limits of esata, and we have to accept it.

cheers

Rainer

factstory
Rainer Wirth
phone_0049-177-2156086
Mac pro 8core
Adobe,FCP,Avid
several raid systems


Return to posts index


Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 5, 2013 at 6:15:28 am

[Rainer Wirth] how is it, that we experience huge speed loss with esata connections in comparison with let's say SAS connections.

If the comparison is SATA vs SAS, SNIA has more than one slideshow that explores the differences that result in better scalability for SAS that go beyond bandwidth. The difference between SAS 3Gbps and 6Gbps, and SATA 3Gbps and 6Gbps isn't just one of bandwidth. There are other meaningful changes to the protocols including command queueing that for certain use cases can make a huge difference. So even when bandwidth saturation isn't the bottleneck, there can be differences simply because of the different command set being used. So while you have not stated exactly what products you were comparing, I'll argue it will be difficult to answer your question because there are a lot of differences possible between SATA and SAS.

But by continuing to ding eSATA specifically, you're implying a meaningful difference between eSATA and SATA which is untrue.

We therefore have decided that SAS and FC are the only solutions for us.

Lots of people have arrived at this conclusion as well, which is why it's good to have SAS as an alternative to SATA. But you keep saying eSATA as if it's different from SATA is sortof annoying.

In the future Thunderbolt might be an answer, but even thunderbolt doesn't reach the speed of a 4x8GB/s FC Raid.

And the specs bear this out. You don't need to test it. But I think you mean 8Gbps or 8GFC rather than 8GB/s. 8GFC is ~1.6GB/s.

In this thread someone is coping with the actual speed limits of esata, and we have to accept it.

No they are not. The bandwidth of the connection is not being saturated even accounting for 8b/10b overhead, on reads let alone writes. The problem is elsewhere.


Return to posts index

Kevin Colber
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 6, 2013 at 8:52:53 pm

FWIW:

I added an ATTO H680 SAS Card with an ESATA Fan Out to my echo express box, which boosted my write speeds into the 195MB/s+ range- someone above mentioned the CalDigit cards and G-Speed Q's didnt like each other- they were right.


Return to posts index

Alex Gerulaitis
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 6, 2013 at 9:28:25 pm

[Kevin Colber] "I added an ATTO H680 SAS Card with an ESATA Fan Out to my echo express box, which boosted my write speeds into the 195MB/s+ range- someone above mentioned the CalDigit cards and G-Speed Q's didnt like each other- they were right."

For a 4-bay eSATA RAID5, that's an awesome speed. Would you have a screenshot by any chance?


Return to posts index


Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 6, 2013 at 11:22:25 pm

What happened to read speeds? 195MB/s writes are still not saturating the SATA 3Gbps link. If the reads are relatively unchanged, then I'd say you don't have an mis-alignment, nor is there excessive RWM.


Return to posts index

Neil Sadwelkar
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 10, 2013 at 4:37:48 pm

This has been my experience too.

G-Speed Q and G-Speed eS drives are picky about the eSATA card you connect with. I've also experienced the same speed bump with the Atto H680 and SAS-4x eSATA cable. But that card-cable combo doesn't work with G-Tech RAIDs if they are 'RAIDed' using the G-Tech SATA card.

It's also very interesting to read all the other comments about your write speed issue.

-----------------------------------
Neil Sadwelkar
neilsadwelkar.blogspot.com
twitter: fcpguru
FCP Editor, Edit systems consultant
Mumbai India


Return to posts index

Alex Gerulaitis
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 10, 2013 at 5:51:09 pm

[Neil Sadwelkar] " I've also experienced the same speed bump with the Atto H680 and SAS-4x eSATA cable."

Very curious about that. Would love to test the Q with a "known good" eSATA card that has been proven to run at 250MB/s (in 3G). Just can't fathom needing a $300 SAS HBA to run the Q at full speeds.


Return to posts index

EricBowen
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 12, 2013 at 3:44:18 pm

The biggest deciding factor to E-Sata Multiplexing bay performance is the E-Sata controller cards. I have tested this extensively from very cheap Marvell Startech controllers to Silicon image controllers and even Highpoint. This is one instance where what you pay for is what you get. This works the same with SAS controllers. Higher end LSI SAS controllers have far greater performance especially with rebuild rates than much cheaper Highpoint SAS controllers. The best E-Sata card that I have seen is Sonnet's. That card will give you close to peak bandwidth performance for your E-Sata multibay regardless of the manufacturer especially on HFS+ arrays. However this is a $250+ expense. The only decent cheaper controllers that I have seen were Marvell and only with the latest firmware and drivers. Keep in mind Caching models on the controllers for command queuing along with the controller firmware initializing with Windows 7 cache policy correctly is what decides this. So the overall here is if you want the 200Mb/s a sec on these 4 and 5 bay E-sata units expect to get a $300 controller card. If 185MB/s Read 150MB/s writes is ok then look at highpoint or Marvell controllers.

Eric-ADK
Tech Manager


Return to posts index

Neil Sadwelkar
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 17, 2013 at 5:49:57 am

I've just run the Blackmagic Speed test on my G-Speed Q 8 TB (RAID5 - 6 TB) with 1.15 TB free. I've connected it to a MacPro 8-core through a Sonnet 2-port SATA card. I'm seeing about 230 MB/sec write and 250 MB/sec read.

-----------------------------------
Neil Sadwelkar
neilsadwelkar.blogspot.com
twitter: fcpguru
FCP Editor, Edit systems consultant
Mumbai India


Return to posts index

Brad Bussé
Re: G Speed Q ESATA Write Speeds in Raid5
on Oct 29, 2013 at 8:18:54 pm

Eric, can you list the exact model of Sonnet card that you're using? Is it the H680 SAS card or one of the SATA cards like the Tempo?


Return to posts index

Chris Murphy
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 21, 2013 at 4:04:37 pm

EricBowen writes: The biggest deciding factor to E-Sata Multiplexing bay performance is the E-Sata controller cards.

I'll speculate that this is due to the difference in switching method used by the controller. Command based causes the controller to communicate discreetly to each drive via the multiplier, one drive at a time and no outstanding commands are permitted. FIS based enables the controller to, in a sense, communicate to all of the drives at the same time, with multiple outstanding requests per drive. Obviously FIS is a lot more complicated, but also a lot more efficient.

What I'm not sure of about the G-Speed Q is how this relates to its internal raid controller, since I don't think it's up to the computer SATA controller to talk to the drives. The G-Speed Q controller presents a logical drive to the computer SATA controller which should just spit out a stream of data to the G-Speed Q. And then it's up to the G-Speed Q controller to divvy that up into data and parity chunks, and deliver them to the proper drives. So I'm a bit puzzled why the controller matters in this case.


Return to posts index

EricBowen
Re: G Speed Q ESATA Write Speeds in Raid5
on Sep 23, 2013 at 3:49:36 pm

I have seen where the Multiplexor controller on the bay slows the performance down with controllers that have better performance on others. Likely due to what you reference. Firmware updates for the bays have changed this. Very similar to what I see with USb3 multibays. However the caching policy is at the E-Sata controller level and that is where the largest performance difference is made. I have also seen this with onboard Intel controllers and raid. They often change their caching policy and the raid 0 performance changes because of that. Sonnet just seems to have a much better caching policy than the other controllers.

Eric-ADK
Tech Manager


Return to posts index

<< PREVIOUS   •   VIEW ALL   •   PRINT   •   NEXT >>
© 2017 CreativeCOW.net All Rights Reserved
[TOP]