D-Link Forums
The Graveyard - Products No Longer Supported => D-Link Storage => DNS-321 => Topic started by: sghayal on January 12, 2009, 02:32:44 PM
-
I just got DNS-321 and found some issues with the speed.
What I am seeing is that transfer speeds are very very sluggish for Gigabit connection. What I am seeing is I get maximum FTP speeds of around 10 to 14 MBPS. (as reported during FTP) but actual speeds are around 450 to 500 KBPS.
This is my configuration.
1) Linux machine with 1 GBPS card
2) D-Link DGS 1008D Gigabit switch
3) DNS-321
4) Windows machine with 100 MBPS card
This are my parameters of comparasion
1) Sending file from windows to Linux (Fastest)
2) Sending file from Linux to DNS-321 (slow)
3) Sending file from Windows to DNS-321 (slow)
During my investigation I found that DNS-321 does not seem to support 1000Base-Tx standard.
If you look at standards supported in document below it does not have 1000Base-Tx
ftp://ftp10.dlink.com/pdfs/products/DNS-321/DNS-321_ds.pdf
This would mean that I have pipeline for 1GBPS but I don't have motor to throw in 1GBPS data.
Can someone clarify this and provide proper guidance.
Thanks,
Sandip Ghayal
-
I do think the basic cause of getting equal speed at 100Base is because there is no support for 1000Base.
Thanks,
Sandip
-
Ok I did proper test on this.
I created close to 7.9 GB file.
I transfered it from Linux machine to my windows machine and Linux machine to DNS-321.
I used Samba mount to mount shared drive from Linux on to windows and mount shared drive from DNS-321 onto Linux.
Both the file transfers took same time.
One thing to note here is when I was doing file transfers to Windows machine from Linux my Linux machine was also servering other network related things like some file transfers to my ATV box.
Thus it makes me conclude that my Linux machine is able to pump out more power then regular 100 MBPS link. Also during the copy to windows I was noticing that the n/w usage on my windows machine was around 85% (meaning entire pipe to window's machine was full).
When I did file transfer from Linux to DNS-321, I ensured that my linux machine was not doing anything apart from doing file transfer and still the rate of file transfer is same as my 100MBPS network.
This clearly proves to me that what I have got from you is 1 GBPS pipe but my pumping motor is 100 MBPS.
So The question is why do I spend $80 more to get this device when its performance essentially is going to be same as device supporting 100MBPS.
I am seriously thinking of returning this device back.
Thanks,
Sandip Ghayal
-
It seems there could be a number of reasons why you're realizing 100Mbps ( lower case b for bits, upper case B for bytes ). Three questions... what kind of network card do you have in your linux box? And does this network card support jumbo frames? Are you using Cat 5e or Cat 6 cables between your linux box and your switch and between your switch and DNS 321?
I ask as all of these could affect your connection speed. I noticed that the D-Link switch you're using supports jumbo frames. I suggest if you have Cat 5e or 6 cables and all devices support jumbo frames you enable it on your Linux box and the DNS-321. Test your speed again and post your results assuming jumbo frames wasn't enabled when you first posted your results.
There's also a setting I've seen, something like "Run at wire speed". If your cables can't support gigabit there's the off chance that your network card or DNS 321, though I'm not sure if the DNS has this capability, is throttling down your connection to a speed the cable(s) can support.
Just a thought. Let us know how it turns out. I enabled jumbo frames on my network, I have an onboard Marvell NIC that supports 9k jumbo frames, a Netgear GS105 that supports 9K frames and the DNS 321 and DNS 323 that also support 9k frames. I've enabled 9k frames on my PC and the DNS' and saw an increase of 3Mbps, from 15Mbps to 18Mbps ( according to a Vista file transfer dialogue window ). Doesn't seem like a huge increase but the time to transfer files dropped significantly.
-
what kind of network card do you have in your linux box? And does this network card support jumbo frames?
Ans>> Yes it does support Jumbo Frames and is configured to use Jumbo frames. I don't have the name of the provider as this box came with the network card.
Are you using Cat 5e or Cat 6 cables between your linux box and your switch and between your switch and DNS 321?
ANS>> Yes
I ask as all of these could affect your connection speed. I noticed that the D-Link switch you're using supports jumbo frames. I suggest if you have Cat 5e or 6 cables and all devices support jumbo frames you enable it on your Linux box and the DNS-321.
ANS>> Results are with Jumbo frames enabled.
Test your speed again and post your results assuming jumbo frames wasn't enabled when you first posted your results.
There's also a setting I've seen, something like "Run at wire speed". If your cables can't support gigabit there's the off chance that your network card or DNS 321, though I'm not sure if the DNS has this capability, is throttling down your connection to a speed the cable(s) can support.
ANS>> This might be bit strange that my cables speed max out exactly at 100 MBPS. But I am using Cat5e/Cat6 standard cables (not crimped by me :-) )
Just a thought. Let us know how it turns out. I enabled jumbo frames on my network, I have an onboard Marvell NIC that supports 9k jumbo frames, a Netgear GS105 that supports 9K frames and the DNS 321 and DNS 323 that also support 9k frames. I've enabled 9k frames on my PC and the DNS' and saw an increase of 3Mbps, from 15Mbps to 18Mbps ( according to a Vista file transfer dialogue window ). Doesn't seem like a huge increase but the time to transfer files dropped significantly
ANS>> And this is where my issue is. We are moving from 100 MPBS to 1000 MPBS signaling speed we should see significant in increase in network speeds almost 5-6 times when using a true Gigabit card.
ANS>> Now this is what I suspect the issue is. I looked at data sheet for DNS321 and DNS323 and I see one significant difference under Standards supported
DNS323 says IEEE 802.3ab 1000Base-T Gigabit where as DNS321 does not mention this standard. Which basically means we have 1GB pipe but have 100 MB pump to fill in. So anything you do you will get same speed as 100MBPS
-
So The question is why do I spend $80 more to get this device when its performance essentially is going to be same as device supporting 100MBPS.
$80 more than what?
www.smallnetbuilder.com has a NAS chart that compares different products. http://www.smallnetbuilder.com/component/option,com_nas/Itemid,190/
If you select the "1000 Mbps 4K Jumbo Aveage Write Performance" benchmark
you will see the DNS-321 can achieve 16.1 MBytes/s which is above the 100Mbits/s (12.5MBytes/s ) theoretical limit of 100M Ethernet.
-
Unfortunately, that's true. Real world gigabit speeds is somewhere above 100Mbps and below 1000Mbps. The rated speed of a network card's a theoretical maximum. For inexpensive consumer hardware I'm not surprised that I'm getting 18MBps, which is 144Mbps, while transferring files. With my 100Mbps network card I would be lucky if I could get 80Mbps. There could be a number of reasons for this, a number of system or network bottlenecks. I'll bet far better network speeds can be obtained if you were to a) transfer your files from a fast data source like a RAM drive b) the network card was connected to a hi speed data bus, PCI-X or PCIe c) the network card had a significant amount of dedicated high speed memory it could use as a buffer.
For testing purposes, one suggestion would be to attach your Linux box directly to your DNS. Setup static IPs and test out your speed once again. Who knows, maybe the switch is the bottlekneck ( sorry d-link ).
-
Folks,
I think we all are missing a point here.
I have 100MBPS and 1000MBPS card and both card have smae thru puts when both of them are flushed with 1000MBPS line on other end, this is really unrealistic scenario if switch is also a Gigabit switch.
The real issue I feel is D-link for some reason beyond my imagination is not suporting 1000-BaseTX standard with DNS-321 though it provides 1000GBPS signaling speed. So that situtation is like D-link provides a pipe that will carry 1000 liters of water /second but the provided pump is only pumping at 100 liters of water / second. Is there a disconnect here?
I am going to try today evening to send file across gigabit network using my laptop which has gigabit card and my linux with gigabit card and see what kind of transfer speeds I get from that.
-
Folks,
I think we all are missing a point here.
I have 100MBPS and 1000MBPS card and both card have smae thru puts when both of them are flushed with 1000MBPS line on other end, this is really unrealistic scenario if switch is also a Gigabit switch.
The real issue I feel is D-link for some reason beyond my imagination is not suporting 1000-BaseTX standard with DNS-321 though it provides 1000GBPS signaling speed. So that situtation is like D-link provides a pipe that will carry 1000 liters of water /second but the provided pump is only pumping at 100 liters of water / second. Is there a disconnect here?
I am going to try today evening to send file across gigabit network using my laptop which has gigabit card and my linux with gigabit card and see what kind of transfer speeds I get from that.
I understand what you are saying exactly, but I doubt D-Link would admit to that fact if it is indeed how the NAS is setup.
-
If you're going to try bypassing the switch definetly let us know how it goes. I was thinking about trying this myself just to see what speeds I could get directly connected to the DNS-321 or 323. Another idea came to mind that may be affecting your speeds, the PCI bus. Is this a PCI gigabit card or an onboard gigabit adapter attached to the PCI bus? The PCI bus could be a serious bottleneck for a gigabit adapter especially if there are a number of other devices in your system causing bus contention issues.
What we're trying to say here is there a number of factors that could affect your gigabit speed. It could be the DNS 321 but it could also be a number of other issues. I hope you figure it out, I'm starting to think my 18MBps isn't nearly enough.
-
...Is there a disconnect here?
Yes, there are so many significant dissimilar characteristics between your two "test cases"
which makes your comparison practically inconclusive.
-
Mig can you specify the reason for disconnect?
Thanks
-
If you're going to try bypassing the switch definetly let us know how it goes. I was thinking about trying this myself just to see what speeds I could get directly connected to the DNS-321 or 323. Another idea came to mind that may be affecting your speeds, the PCI bus. Is this a PCI gigabit card or an onboard gigabit adapter attached to the PCI bus? The PCI bus could be a serious bottleneck for a gigabit adapter especially if there are a number of other devices in your system causing bus contention issues.
What we're trying to say here is there a number of factors that could affect your gigabit speed. It could be the DNS 321 but it could also be a number of other issues. I hope you figure it out, I'm starting to think my 18MBps isn't nearly enough.
Thats something I want to try but little bit scared of spoling my setup on Linux. I have found some issue on my Linux machine that if not connected to network properly causes my box to hang or reply really slow so don't want to touch that box much at this time as it is very critical component in my home network (running FTA TV :-) ).
I agree with yoru comments about PCI card and going to find it out but still I would hope slightly higher speed of transfer on 1GB line then 100MB line and strangely I am not seeing that.
-
If I could just chime in here ...
If you can transfer data at anything over 100 mbps, you're getting gigabit speeds - a lot of people do not seem to recognize that the transfer rate observed when copying files from one system to another is only as fast as the slowest link in the chain, which starts with a hard drive at one end and finishes with a hard drive at the other - the network may not be the bottleneck.
Back in the days of 10 mbps ethernet, 3Com's "parallel tasking" line of cards was the only thing that could approach wire speed, and to make that happen I had to load one card in a server with three data streams from client PCs using a memory-to-memory test procedure that 3Com outlined.
In those days the average throughput on a 10 mbps ethernet card was <3 mbps.
When 100 mps ethernet came along, I ran those same tests and got throughput in the vicinity of 25~35 mpbs and gigabit ethernet is no different - depending on the hardware - low end hardware ~170 mbps, high end hardware ~800 mbps.
I don't have a DNS-321 and so can not provide figures for it, but the older DNS-323 can deliver ~400 mbps - memory-to-memory, this does not include the transfer to/from the disks, with a disk-disk transfer I can hit 28MBPS on a read, or 224~280 mbps.
The 15~18 MB/s reported by the second poster is indicative of giagbit speeds.
-
Mig can you specify the reason for disconnect?
You have preformed a completely uncontrolled test, therefore no reasonable conclusion can be reached.
You have chosen to test your DNS-321 with two VERY different computers:
different OS, different network cards, different motherboards, different filesystems.
And from that you conclude the DSN-321 does not support 1000Mbps network?!! :o
What I think you should be trying to figure out is, why your linux box can't achieve the
transfer rate that smallnetbuilder.com (and others) are able to achieve, instead of making
claims that the DNS-321 is flawed.
As Freeman posted:
What we're trying to say here is there [are] a number of factors that could affect your gigabit speed. It could be the DNS 321 but it could also be a number of other issues. [...]
By the way, I'm not saying your tests are not valid. These are your "real world" operating
conditions, regardless of what transfer rates anyone else is able to achieve. If you are not
able to achieve satisfactory transfer speeds in your environment, then maybe this is not the
right NAS for you.
-
And from that you conclude the DSN-321 does not support 1000Mbps network?!! :o
Please answer my simple question, why does not DNS-321 to not support 1000-Tx standard?
Thanks
-
The DNS-321 does support 1000-Tx standard.
I think the reason you are not achieving faster transfer rates has more
to do with your computer and your network than the nic in the DNS-321.
The best rates you should get (from empirical test results www.smallnetbuilder.com)
is around 16MBytes/sec.
Try this experiment (if you have the hardware), put a gigabit nic in your windows
computer and run this NAS performance tester http://www.808.dk/?code-csharp-nas-performance
-
You have preformed a completely uncontrolled test, therefore no reasonable conclusion can be reached.
You have chosen to test your DNS-321 with two VERY different computers:
different OS, different network cards, different motherboards, different filesystems.
And from that you conclude the DSN-321 does not support 1000Mbps network?!! :o
What I think you should be trying to figure out is, why your linux box can't achieve the
transfer rate that smallnetbuilder.com (and others) are able to achieve, instead of making
claims that the DNS-321 is flawed.
Well it not totally uncontrolled environment. This is totally controlled environment where I have 2 d-link devices (1) switch and (2) NAS and from what I am seeing is the performance is matching that of another device which only had 100MBPS line. Offcourse your doubt is valid that Linux might not be able to pump at higher speeds but if you look at my experiments during windows tests, I have loaded linux box with other network related functions too, like transfering same ISO image to Apple TV box and watching live movie at the same time. And there was no difference in transfer speeds with my windows machine with windows network piple totally full at at 80-85%. Thus I am concluding that my Linux machine is able to transfer at higher then 100 MBPS rate.
The question that still remains unanswered is why does DNS-321 not support 1000-Tx standard. Should not it be implied that when you support 1000mbps signalling you should also support 1000-tx standard.
Thanks
-
The DNS-321 does support 1000-Tx standard.
Not to be offensive but how can you claim DNS-321 supports 1000-tx (technical name for the standard is IEEE 802.3ab 1000Base-T Gigabit) when Dlink folks don't claim to support it.
Look at the links below and see if you can find support for 1000-tx standard.
ftp://ftp10.dlink.com/pdfs/products/DNS-321/DNS-321_ds.pdf
-
sghyal
How about you read the thread, and look at some of the links that other users of the DNS-321 have contributed.
How can you insist that it provides only 100mbps performance when ...
a) in the first post YOU claim 10-14 MBPS - that requires 80-112 mbps - which is faster than 100mbps.
b) the fourth response indicates a jump from 15-18 mbps with the use of jumbo frame - 15-18 mbps is 120-144 mbps, again faster than 100 mbps, and jumbo frame does not run on 100 mbps networks.
c) the sixth post links to a review claiming 16.1 MBytes/s - or - 128.8 mbps, again over 100 mbps.
The suggestion has been made that your network infrastructure is what may be limiting you to low performance, you apparently choose to ignore this - other folks are obviously getting >100 mbps out of their DNS-321 - why can't you?
Oh sorry - you are too - based on your first post.
-
I think everyone is missing his main point because honestly, most of us would not consider it applicable...
In THIS PDF ftp://ftp10.dlink.com/pdfs/products/DNS-321/DNS-321_ds.pdf , on page 2, under "Standards" it does NOT list 1000Base-TX. It is the lack of this text in the PDF that sghyal is basing his assumptions from. Since the device does indeed connect at 10/100/1000Mbit speeds, I would assume this was simply a mistake and D-Link left it out of the PDF.
That's it, period, end of story...
-
I read the PDF when he first linked to it - and I don't think that's his main point - I think it's the straw he's grasping at (as in a drowning man will ...)
His problem, based on the first three posts is that he's dissatisfied with the throughput he's getting - he's been repeatedly told that a gigabit network will not necessarily deliver gigabit throughput and also that anything over 100 mbps - which he is apparently getting - can be considered gigabit speeds.
-
I read the PDF when he first linked to it - and I don't think that's his main point - I think it's the straw he's grasping at (as in a drowning man will ...)
His problem, based on the first three posts is that he's dissatisfied with the throughput he's getting - he's been repeatedly told that a gigabit network will not necessarily deliver gigabit throughput and also that anything over 100 mbps - which he is apparently getting - can be considered gigabit speeds.
Fact remains, until D-Link comes in and says "Yes, we support 1000BaseTX, the PDF was a misprint" he's going to be convinced it's the unit and not his network.
-
I really don't think D-Link admitting the PDF has a mistake and/or them correcting it will change how he feels - so it supports 1000BaseTX - all that does is confirm the 1000 litre/second pipe ...
D-link provides a pipe that will carry 1000 liters of water /second but the provided pump is only pumping at 100 liters of water / second
He's concerned about the 100 litre/second pump and realistically speaking - it's what - a 300 litre/second pump?
The DNS-321 cannot deliver anywhere near gigabit/second speeds, many gigabit equipped devices can't.
-
I'm seeing very similiar results as sghayal.
I decided to upgrade to a gigabit network (finally) in hopes to speed up backups and copies to my NAS.
I installed a dlink DGS-2208 8 port gigabit switch.
Before installing my new gigabit nic card, I decided to run some quick test. I took a 4.35 GB (4,681,455,616 bytes) iso and copied it from my dns-321 to my desktop over wireless. I let the copy go on for about a minute and recorded the stats:
2.3 MB/s
32 minutes
Then I pulled out the wireless card and plugged in a network cable to the onboard nic and ran the test again and got the following (running at 100 on the onboard nic):
7.5-7.7 MB/s
8:50 minutes
I then put in my new gigabit lan card (I'll admit it's a cheap one from Trendnet model teg-pcitxr) and ran the test again and only saw a slight improvement.
11.1 MB/s
6:50 minutes
It's faster, yes, but nowhere near what I expected. I would never assume the gigabit nic would run 10x faster than a 10/100 nic running at 100, but 3-4x is what others are seeing. I'm only see an increase in 1.5x so it got me wondering if my switch was running at 100MB/s in mixed mode. I confirmed this switch does not pull down all ports in mixed mode to lowest common denominator speed, so I assume I have issues with the dns-321.
After seeing sghayal's post, I believe that although the dsn-321 does support the gigabit signaling, it cannot maintain a gigabit bandwidth rate.
I'd like to test this on a dns-323 which I have, but it's boxed up with no drives to do testing with.
Yes, I know I am not testing this in a controlled environment. It's controlled enough for me and it should have been at the very least be significantly faster (when I say significant, 2x would do :) ).
Gary
-
Ran another test...
After copying the iso from the nas (DNS-321) to the desktop, I copied the same file from the desktop to the laptop. Both machines have gigabit NICs.
From DNS-321 to Desktop on a gigabit network I get 11.1 MB/s (6:50 - time to copy the file)
From desktop to dell laptop (both have gigabit nics on a gigabit network) I get a whooping 23-26MB/s with a copy time of about 3:15. I copied the file back from the laptop to the desktop and I get similiar performance numbers.
I don't have jumbo frames set, but that shouldn't be a major issue. I have a feeling my dns-321 isn't performing up to gigabit performance levels.
On my 10/100 nic running at 100 this is a 3x (almost 4x) increase in performance comparison. I'm happy with this inexpensive nic card. I'm not so happy with the gigabit performance on the dns-321 (lets hope this isn't an issue with the dns-323 which I also have :( ).
Gary
-
If it's > 100 mbps, it's gigabit performance - you may not be happy with it, but it's gigabit performance.
The DNS-323 will deliver roughly the same speeds without jumbo frame, and with jumbo enabled you'll probably see 30~35 MB/sec.
Now let me ask this - and more to open your eyes than get an answer ...
Where is the bottleneck? Is it the network infrastructure itself (the cables, router & switches), is it the DNS-321 (the disks, the processor, the memory, the NIC), is it the other computer (the disks, the processor, the memory, the NIC)
I'm transferring data from my IBM server to my DNS-323 - a backup - throughput, according to my network switch is 3,285 kbit/sec - :O
Just in case you're curious the NIC in a DNS-323, which I believe is the same in the DNS-321 is capable of 400 mbit/sec transfers - roughly 50 MByte/sec - so there's a pretty big difference between what it can pass and what you have measured - might be an idea to find out why.
-
If it's > 100 mbps, it's gigabit performance - you may not be happy with it, but it's gigabit performance.
The DNS-323 will deliver roughly the same speeds without jumbo frame, and with jumbo enabled you'll probably see 30~35 MB/sec.
Now let me ask this - and more to open your eyes than get an answer ...
Where is the bottleneck? Is it the network infrastructure itself (the cables, router & switches), is it the DNS-321 (the disks, the processor, the memory, the NIC), is it the other computer (the disks, the processor, the memory, the NIC)
I'm transferring data from my IBM server to my DNS-323 - a backup - throughput, according to my network switch is 3,285 kbit/sec - :O
Just in case you're curious the NIC in a DNS-323, which I believe is the same in the DNS-321 is capable of 400 mbit/sec transfers - roughly 50 MByte/sec - so there's a pretty big difference between what it can pass and what you have measured - might be an idea to find out why.
Thanks for your reply.
Ok, if I'm running a tad over 100 mbps, then it's gigabit performance? If that's how it's definate, then I won't argue that point, but that is damn poor performance, don't you agree? To me it's like saying I have a Porsche twin turbo... a super car, but for some reason I can't go over 60mph when I floor it.
Now, you're saying that I'm having issues elsewhere and I should be able to get "roughly 50 MB/s". When I achieved over 20MB/s between two computers, I was using the same switch, the same type of wires, yet I'm only getting around 10MB/s with the dns-321. I'll be happy to be around 20MB/s with the dns-321. If i have a bottle neck what do you suggest I try? It's a simple network here. I have two gigabit switches and I can swap them out but why? They are both brand spanking new and the switch indicates I'm in gigabit mode. The cables are cat 5e. I can swap all these if you like, but the question is more than 1 forum member is experiencing the same issue. I just verified last night I can get a transfer rate of about 3-4x faster than my computers equipped with 100mbps NICs. I just can't do it with the dns-321.
You're saying with jumbo frames I should be seeing 30-35Mb/sec. That I haven't tried yet, but a forum member here said that jumbo frame was good for 2-3MB/sec only. If this is case, then I currently should be running at 27-28MB/sec and I'm not anywhere near that ballpark.
What size jumbo frame are you using? Unfortunately the disadvantage of the new gigabit nic card I have is that jumbo frames are limited to about 7K whereas most gigabit nics allow 9K frames.
Gary
-
I'm transferring data from my IBM server to my DNS-323 - a backup - throughput, according to my network switch is 3,285 kbit/sec - :O
3,285 kbit/sec = 3.285mb/sec is roughly .41 MB/sec? is that right?
-
Ok it looks like my NAS is running as it should. Our gigabit NAS just doesn't perform at the gigabit level although it will RUN on a gigabit network, but I guess for the price what are we to expect.
http://www.smallnetbuilder.com/content/view/30521/75/1/3/
http://forums.dlink.com/index.php?topic=3782.msg21604#msg21604
-
Gary,
I actually started posting a response based on a "super car" analogy, but didn't - I didn't think you'd appreciate it ;) but since you bring it up ...
I've got two NAS here, the first is a Taiwanese made Encore Electronics unit, the second is my D-Link DNS-323 - the Encore Electronics is a single disk linux based NAS with a 10/100 mbps port, it's good for ~3.5MByte/sec tops, the D-Link on 10/100 will easily triple that 9~10MByte/sec, on gigabit maybe 18~20MByte/sec, and on gigabit with jumbo frame (9000 byte) another 50%, 27~30MByte/sec.
Let's turn these into cars - that Encore would be your Geo Metro and the D-Link maybe a Toyota Turbo Supra (the gigabit port is the turbo charger) - they can both run on the I75 and the Geo will struggle to make the speed limit, whilst the Supra will do it with ease - now let's put them on the Autobahn in Germany (where there is no speed limit), the Supra looks reasonable maxed out at 150 mph until some dude comes along in his Bugati Veyron and blows by at 250.
So ...
What sort of money did you spend - Geo Metro money? Toyota Supra money? Bugati Veyron money? A Veyron by the way retails at USD$1.5M
-
Read very carefully - I didn't say you could get 50MB/sec from a DNS-323 (or DNS-321), I said the NIC can handle that, and that is based on personal testing. Unfortunately the NIC is not the bottleneck so if you transfer a file from one computer to the DNS-323 (or vice versa) you're not going to see 50MB/sec, you're going to get the maximum speed that the bottleneck can pass - whatever that bottleneck is or wherever it happens to be.
By the way - yes - the throughput when backing up my IDM server is to the order of .5MB/sec and trust me that IBM cost me more than the DNS-323 did, and is capable of reading/writing a lot faster (the NIC there is good for 800 mbits/sec or 100MByte/sec.
Confused?
I did say the idea was to open your eyes, didn't I.
Transferring a data file from one disk to another across a network is really not that simple a process.
You start by reading data off of the first disk, passing it through a disk interface across an expansion bus and through a network interface, all in one device - then out on to the network, which may be as simple as a cross-over cable, or may be complex and include switches and routers and wireless - and then into the second device, through a network interface, through an expansion bus, through a disk interface and finally write to the disk.
Your transfer speed will be as fast as the slowest of the items mentioned.
Reading a single large contiguous file from a disk is fairly quick, but if that disk is fragmented and the head has to jump around to retrieve the data - a block here, a block there, the transfer speed is going to take a major nose dive, and the same thing happens if the data being transferred is a large number of smaller files (now you know why my backup takes as long as it does).
Where is the bottleneck? It's hard to tell, but it's in the DNS-323/321 and probably the processor, which has to calculate checksums etc. during the transfer process.
Thanks for your reply.
Ok, if I'm running a tad over 100 mbps, then it's gigabit performance? If that's how it's definate, then I won't argue that point, but that is damn poor performance, don't you agree? To me it's like saying I have a Porsche twin turbo... a super car, but for some reason I can't go over 60mph when I floor it.
Now, you're saying that I'm having issues elsewhere and I should be able to get "roughly 50 MB/s". When I achieved over 20MB/s between two computers, I was using the same switch, the same type of wires, yet I'm only getting around 10MB/s with the dns-321. I'll be happy to be around 20MB/s with the dns-321. If i have a bottle neck what do you suggest I try? It's a simple network here. I have two gigabit switches and I can swap them out but why? They are both brand spanking new and the switch indicates I'm in gigabit mode. The cables are cat 5e. I can swap all these if you like, but the question is more than 1 forum member is experiencing the same issue. I just verified last night I can get a transfer rate of about 3-4x faster than my computers equipped with 100mbps NICs. I just can't do it with the dns-321.
You're saying with jumbo frames I should be seeing 30-35Mb/sec. That I haven't tried yet, but a forum member here said that jumbo frame was good for 2-3MB/sec only. If this is case, then I currently should be running at 27-28MB/sec and I'm not anywhere near that ballpark.
What size jumbo frame are you using? Unfortunately the disadvantage of the new gigabit nic card I have is that jumbo frames are limited to about 7K whereas most gigabit nics allow 9K frames.
Gary
-
Fordem,
Using your analogy with our dns-321/323, then there is no way your veyron will go over 60mph :).
I like your supra analogy - I have a real old one that puts out 450HP :), but if it was like my dns-321 it won't go faster than a geo.
There is no way our NAS will hit 27-30mb/sec for a big file - over 1 gig. Correct me if I am wrong. Even these guys
http://www.smallnetbuilder.com/content/view/30521/75/1/3/
couldn't get over 20MB/sec with large files. They're stuck around 15-16MB/s. So unless you run some sort of military spec dns-321 on a mil network, I don't see how you can get that throughput. I'm not saying it's impossible, I just can't see how you can do it.
Are you running vista? If so can you do a little experiment for me. Find a large file on your NAS drive. Perhaps an DVD iso image. Drag and drop that from your NAS drive to your desktop. When the copying dialog comes up tell me what it says for the transfer rate after it settles down (30-60 sec). I'm just curious what your transfer rate is.
-
I did read that carefully. You didn't say our NAS will go 50MB/s. You said the NIC on these nas is roughly capable of 50MB/s:
Just in case you're curious the NIC in a DNS-323, which I believe is the same in the DNS-321 is capable of 400 mbit/sec transfers - roughly 50 MByte/sec - so there's a pretty big difference between what it can pass and what you have measured - might be an idea to find out why.
Read very carefully - I didn't say you could get 50MB/sec from a DNS-323 (or DNS-321), I said the NIC can handle that, and that is based on personal testing. Unfortunately the NIC is not the bottleneck so if you transfer a file from one computer to the DNS-323 (or vice versa) you're not going to see 50MB/sec, you're going to get the maximum speed that the bottleneck can pass - whatever that bottleneck is or wherever it happens to be.
By the way - yes - the throughput when backing up my IDM server is to the order of .5MB/sec and trust me that IBM cost me more than the DNS-323 did, and is capable of reading/writing a lot faster (the NIC there is good for 800 mbits/sec or 100MByte/sec.
Confused?
I did say the idea was to open your eyes, didn't I.
Transferring a data file from one disk to another across a network is really not that simple a process.
You start by reading data off of the first disk, passing it through a disk interface across an expansion bus and through a network interface, all in one device - then out on to the network, which may be as simple as a cross-over cable, or may be complex and include switches and routers and wireless - and then into the second device, through a network interface, through an expansion bus, through a disk interface and finally write to the disk.
Your transfer speed will be as fast as the slowest of the items mentioned.
Reading a single large contiguous file from a disk is fairly quick, but if that disk is fragmented and the head has to jump around to retrieve the data - a block here, a block there, the transfer speed is going to take a major nose dive, and the same thing happens if the data being transferred is a large number of smaller files (now you know why my backup takes as long as it does).
Where is the bottleneck? It's hard to tell, but it's in the DNS-323/321 and probably the processor, which has to calculate checksums etc. during the transfer process.
-
Gary,
I don't know if this is going to work - I've never used imageshack before - anyway, let's see how it goes.
(http://img105.imageshack.us/img105/5527/dns323la4.th.jpg) (http://img105.imageshack.us/my.php?image=dns323la4.jpg)
If it works, it's going to need a bit of an explanation.
This is a composite of three screen captures assembled in MS Paint, because I couldn't capture all three images in one shot.
The background is an SNMP graphing tool called PRTG (www.paessler.com), it reads the bandwidth figures from my network switches, the one that's open is the graph from the switch port that feeds my DNS-323.
Directly below that is the output from a little utility called NASTester - if you search the DNS-323 forum, you should find a link to another site that you can download it from. Basically what it does is create a test file, of a size that you choose (up to 2GB) on a local drive and then measure the time it takes to transfer it across the network to a mapped drive, it then calculates the average transfer speed and displays it - you can tell it how many iterations and it will loop as required and list each run and then give you the average, and then it reverses the two end points and repeats, so you get both write speeds & read speeds.
To the left of that is the Windows task manager displaying the network tab and the bandwidth that Windows is measuring.
From these three you can see here that with a 2GB file size, I can read approximately 250 mbit/sec from my DNS-323, achieving an average of 27.96 Mbyte/sec over two runs. I have seen transfers averaging in excess of 30 Mbyte/sec and instantaneous peaks (reported by DUMeter) exceeding 35 Myte/sec.
There is nothing "mil spec" about my DNS-323 or my network - in fact, it's all entry level off the shelf stuff - a DNS-323 Hardware Rev A, firmware 1.06, with a pair of Seagate 250GB, Barracuda 7200.9 desktop SATA drives in a RAID1 configuration, an IBM xSeries 206 server (this was entry level three years ago, cost me all of $746),a desktop 3.0 GHz Pentium IV, with 1GB RAM, two 250GB desktop SATA drives - 1 x Seagate Barracuda 7200.9, 1 x Maxtor QuickView 6L250S0 - in a RAID1 configuration on the integrated Intel ICH SATA controller, integrated Intel PRO/1000CT network interface, running MS Windows Server 2003 standard edition. These are connected by CAT5 (yes CAT5, not CAT5e, not CAT6) cables to a Netgear GS108T 8-port SmartSwitch - jumbo frame is enabled and set for a 9000 byte/frame size.
I'm sorry - I don't have a Vista system on hand - I'm still running XP Pro, a couple of my kids have Vista laptops, but nothing with a gigabit port, so I don't know that borrowing one of those would serve a purpose.
By the way - this is the same exact hardware/software/network that chugs along at 0.5 MB/sec doing a backup, so file size does have a significant impact on throughput.
-
Wow, I come back and the discussion continues! :o
I think the technical back and forth can go on for quite some time but one thing can be agreed upon.
That the DNS 321 can transfer at gigabit speeds although at the lowest end if only briefly in some cases.
This is really kind of beside the point, though, isn't it? It's faster than the fastest 100Mbps network adapter. It's also as fast as it's going to get on a DNS 321 with its limited hardware, its an immensely underpowered little computer which is sufficient for a NAS device. Low powered CPU, small amount of memory...There are so many possible bottlenecks in there that it really isn't worth the effort arguing about it. It's not much competition comparing it to a modern PC or laptop that has far beefier hardware.
What it boils down to is price versus performance and in my opinion, other than the current buggy firmware which we all hope D-link will fix very soon it's currently the best NAS in its class. Try to find a NAS from another manufacturer that offers the same features at this price point ( D-Link, fix the bugs please! ).
I like the car analogy so I'll try my own, bear with me. I think it would be better to introduce a Honda Civic to represent a 100 Mbps card and the Porsche can continue to represent the 1000 Mbps card. Except you bought the Porsche for the cost of a Honda Civic ... or a geo metro. Now you're driving it at 140 mph and you think, damn, why can't I go at its maximum speed, 180 mph?! The Honda Civic, BTW, has a max of 120... it's all relative... and the Porsche is still fast.
Just out of curiosity, are you comparing the DNS 321 to the performance of any other NAS'? I'm just wondering if there are other gigabit NAS' with gigabit cards, at the time I bought my DNS 323, my first NAS, it was the only one with a gigabit port at its price.
BTW, I set my jumbo frames at 4k, I read in a review that it was the DNS 321 sweet spot and there was no real difference at 9k.
-
Freeman
Whilst I agree that these represent "the low end" of gigabit speeds, there is nothing brief about it - the duration of the time that I can hit the ~28 MByte/sec transfers shown is limited only by the size of the files and the available space - the largest file I've ever had on my DNS-323 was ~60GB, and it will sustain the rates shown.
-
fordem:
Running the NAS performance tester, I'm getting the following:
Dell D620 Laptop with gigabit docking station:
Avg (W): 12MB/sec
Avg (R): 9.41 MB/sec
Dell Inspiron530 with an inexpensive gigabit NIC:
Avg (W): 13.93 MB/sec
Avg (R): 9.47 MB/sec
My WRITES are much faster than my READS.
I just changed my network this past weekend and I go through two gigabit switches (both dlink DGS-2208) to get to my NAS. Do you think that will slow my throughput?
If I start using jumbo frames, I think I'll be in your ballpark in the WRITEs, but your READs are amazing. At first I thought perhaps the high performance writes was due to caching but that didn't make any sense.
About the "milspec" stuff, I was only being sarcastic :). You are running the intel PRO NIC and they do have rave reviews so I wonder if that combined with the jumbo packets are the two major reasons why you network is faster than mine :(.
As a newbie to gigabit networks, for me to enable jumbo frames, do I need to do anything else to my computers or is that automagically supported by the gigabit nics? Will this affect 100mbps networks with jumbo frames enabled or only gigabit networks? I have my network speed set to "auto" on my NAS.
I'll google jumbo frames right now.
Freeman:
Yes, i agree that performancewise we are at the _VERY_ low end of the gigabit speed.
-
Gary
I'm guessing that your laptop hard drives are the lower performance 5400 rpm ones - the fact that your write sppeds are higher than your read suggests that your laptop cannot write as quicky as it reads - you might also want to try defragging your disks ;) it can make a major difference.
Jumbo frame only works on gigabit networks, you need network cards and a network switch that support jumbo frame and you'll need to specifically enable jumbo frame and in most cases set the frame size. Some switches it'e either on or off, with no frame size settings, and some low end switches that support jumbo frame have it permanently enabled.
The golden rule is that the entire network path between the two devices using jumbo frame MUST support jumbo frame AND the frame size to be used. You may see this statement corrupted to say the entire network, but that is incorrect.
You can mix gigabit with jumbo frame, gigabit without jumbo frame, 100 mbit, 10 mbit and wireless all in the same network and have
-
I certainly appreciate fordem's input and the measurements he's taken time to capture. One major fallacy with his assertions is he's measuring with the DNS-323, while this thread is about the DNS-321. While they are similar in performance, they use different chips and have different firmware. Controlled environs...
I have to agree with garyhgaryh here. I've benchmarked the '321 on my Gb network and it tops out at 16 MBps (MegaBytes p/s). No way it can hit 27 MBps when transferring to its internal HD. The network isn't the bottleneck; I and others who've posted in this thread have performed numerous tests between other systems using the same network that the '321 sits on, and results have proven that the network can handle substantially more traffic than the '321 can push. The limitation is the '321. For its price I can't complain too much :)
An oddity I've noticed is that jumbo frames reduce performance significantly. I have a D-Link Gb switch which claims jumbo frame support, and my PC's NIC supports jumbo frames. I've tried all settings of jumbo frame sizes on the '321 and my PC. None offer any improvement in speed, and most all jf settings reduce performance 10-30%. This looks to me like an issue with the 1.01 firmware which added a UI selection for jumbo frames. Hopefully they will improve jf performance in future firmware releases.
-
Hey peas
I'm not sure I would describe it as a fallacy - I made it clear from the outset that the figures came from a DNS-323 and the measurements were originally intended to illustrate the point I was making, that anything over the 12.5MB/sec performance possible with a 100 mbps network should be considered as gigabit performance, which I believe is apparent if you take the time to read the thread in its entirety
I'd also question the rationale behind calling it a fallacy when you yourself say ...
While they are similar in performance...
- isn't that exactly what my point? I never said they were identical, just similar.
Last - I don't know if you saw this (http://forums.dlink.com/index.php?topic=3280.msg22186#msg22186), but here's someone whose jumbo frame experience on a DNS-321 doesn't match yours - I'd say, if you can't get an increase in throughput when you enable jumbo frame, any increase at all, you're not doing something right. My first foray into jumbo frame about eighteen months back was equally disappointing, but I've since had much better results, using exactly the same equipment - I never did figure out what I was doing wrong.
-
I'd also question the rationale behind calling it a fallacy when you yourself say ... - isn't that exactly what my point? I never said they were identical, just similar.
I'd say we're splitting hairs here. You're throwing around numbers like 27 MBps while people are achieving ~15 MBps. It adds to the confusion at the very least. Instead of pursuing the "I'm right you're wrong" angle, we'd all be better served by staying on topic and understanding the issue at hand.
I agree that any rates above 100 Mbps (12.5 MBps) indicate that the link speed is 1 Gbps and that the '321 is trying as hard as it can to fill that bandwidth. It just has an underpowered CPU; no sense bashing people's network setups. I know a bit about networks, from the TCP layer to IP to raw Ethernet frames. I've modeled Ethernet's CSMA/CD architecture and worked on TCP termination. My network is kosher. I can transfer between 2 PCs on my GbE network at approx 350 Mbps with jumbo frames. That's using Windows' inefficient copy-paste GUI. I don't see an increase in throughput with '321 jumbo frames. In fact I get a slowdown. That's a pretty clear indication something is wrong with the 321's jumbo frame support. Sure there might be one particular config where jumbo frames might work well, but neither I nor numerous others have witnessed it. Mind you, I'm using one of D-Link's own products for the GbE switch (DGS-2208). I'm disappointed that D-Link hasn't tested more thoroughly with at least their own products. Regardless, I've turned off jumbo frames and am satisfied with the '321 as it is. I hope that D-Link will optimize the 321's performance, but it serves its purpose well in the here and now.
-
I believe the DNS 321 is VERY limited in transfer speeds.
I have 2 machines running windows 7 and i can get sustained 70-80MBs but 15MBs max transferring to or from my Dlink NAS..
ill be selling it and building a FreeNAS machine with aggregated Gigabit nics.
-
ill be selling it and building a FreeNAS machine with aggregated Gigabit nics.
I considered that, but decided to try the Synology DS209, and my benchmark speeds came up to very usable numbers. :)