When I first started using a cache drive I did not have any plugins installed and I would get transfer speeds at about 50 MB/s. I have since added an apps cache only folder and plugins live on the cache drive. Now when I transfer files I get speeds around 20-30 MB/s. I am assuming this is normal because the drive is doing multiple things at once? Also, the cache drive is a spinning drive, do SSD drives behave the same way (ie. yield slower transfer speeds if there is an app folder on them)?

Are all of your shares cache-enabled?    Write speeds in the 20-30MB/s range sound like writes to the protected array ... NOT to the cache drive.

 

Unless the cache-only share is VERY busy, that would not account for the slow-down you're seeing.  But writes that are bypassing the cache and going straight to the protected array certainly would.

 

I double checked and all of the shares are set to "Yes" to use cache disk. The cache drive is a 2.5" hard drive if that would make a difference?

So the shares are all set to use the cache; and all the writes you're talking about are to one of those shares -- right?

 

If so, I don't understand the slowdown.  A 2.5" drive will typically have somewhat slower write speeds than 3.5" units; but even the "high" speed you noted (50MB/s) is relatively slow for a modern hard drive.  But a 320GB WD Black is a good drive that should easily exceed the speeds you're noting.

 

What kind of transfer speed do you get across your network if you copy a large (1GB) file from one of your data drives to your PC?

 

 

By the way, you asked a question r.e. SSDs => NO, they do not slow down like rotating platters do when there are multiple accesses.  That characteristic is due to the need to move disk heads and wait for the data to spin in place ... neither of those events happen with SSDs      The 15-20ms typical access time for a spinning 2.5" drive is about 0.01ms with an SSD (it does take some time for the NAND gate addressing to settle).

 

hmmm 1.5 GB file copies at 44 MB/s from computer to computer. That is rather slow correct?

Yes. On a gigabit network you should be getting 100MB/s up to about 120MB/s assuming the underlying disk subsystems in the computers are capable of those speeds.

 

EDIT: Where did you copy the file to/from?

I copied the file from my personal computer desktop to the media pc desktop (the path to get to them includes : personal computer -> switch -> switch -> media pc). I use all Cat 6 cables and the switches are suppose to be GB switches.

With your drives I wouldn't be expecting anything over 100MB/s, but I would expect more than 44MB/s.

 

Your sig doesn't show the specific model numbers of your drives; but I'd guess that you have 333GB and 500GB platters on those drives ... so the areal density is 1/3 to 1/2 that of a modern 1TB/platter drive.

 

That translates to slower sustained data rates.  If, for example, you were copying a file off the inner cylinders of a 333GB/platter drive, then 44MB/s is probably about what you should see.

 

It also depends on the specific drive you were copying TO ... and also whether or not your destination system is set for AHCI mode.

 

 

While the speeds between your PCs are instructive, they're not what I wanted to see.    The relatively low speed is likely for the same reasons I noted above (low areal density, and possibly copying to and/or from inner cylinders).

 

Copy a large file from one of the UNRAID data drives to your PC.  Preferably from the Toshiba 2TB (likely a 500GB/platter unit, although newer models are 1TB/platter) to a relatively empty drive on a PC.

 

I copied a 4.6 GB file from the 2TB Toshiba (Toshiba 3.5" HDD DT01ACA200) drive to my laptop which is about 1/3 used. TeraCopy reports 37 MB/s.

 

 

 

 

Or better yet use something like LAN Speed Test to remove the PC drive from the equation altogether. You would then effectively be testing unRaid and the network only.

 

http://www.totusoft.com/lanspeed1.html

TeraCopy is notoriously unreliable at reporting speed.    Is your laptop Windows 7 or 8?  If so, just use the native Copy command ... it will show the copy speed and is much more reliable than TeraCopy (not a more reliable copier ... but a more reliable speed indication).

 

 

Laptop is Windows 7, I removed TeraCopy and ran the same test and native Copy was between 70-80 MB/s to copy a file from the server to my desktop. What are your thoughts?

 

I downloaded Lan Speed Test but it is telling me I have to register to get the Mbps. EDIT: I tried Lan Speed Test (lite) and tested a 1000 MB file and the results are 306 Mbps Write and 786 Mbps Read.

So that's 38MB/s write and 98MB/s read. Completely normal speeds. 

 

Run the same LAN speed test to the cache drive.

results to the root of the cache drive 320 Mbps (40MB/s) Write and 797 Mbps (99 MB/s) Read.

The writes should definitely be faster unless that drive is close to full. Benchmarks of that drive show max read/write speeds of about 90/80 on the very outer cylinders, with an average across the drive of 71/66. Even the innermost cylinders get mid-40's.

 

Also, as Gary stated before, plugins running on the cache drive could slow down transfer rates a bit but it shouldn't be this much. Just out of curiosity, what plugins are you running from the cache drive?

 

Have you pulled a smart report on that drive?

I ran a smart test and the drive did start making a clicking noise when I ran the test. I know that hard drives and clicking noises are typically bad (although the test looks fine to me...) The drive currently has 80 GB of used space out of 320 GB. I updated my signature with the plugins I currently use.

 

Smart Test

smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build)
Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Western Digital Scorpio Black
Device Model:     WDC WD3200BEKT-00F3T0
Serial Number:    WD-WXM209SM0781
LU WWN Device Id: 5 0014ee 2028374ca
Firmware Version: 11.01A11
User Capacity:    320,072,933,376 bytes [320 GB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    7200 rpm
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ATA8-ACS (minor revision not indicated)
SATA Version is:  SATA 2.5, 3.0 Gb/s
Local Time is:    Fri Dec 27 14:52:04 2013 CST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x00)	Offline data collection activity
				was never started.
				Auto Offline Data Collection: Disabled.
Self-test execution status:      (   0)	The previous self-test routine completed
				without error or no self-test has ever 
				been run.
Total time to complete Offline 
data collection: 		( 7560) seconds.
Offline data collection
capabilities: 			 (0x7b) SMART execute Offline immediate.
				Auto Offline data collection on/off support.
				Suspend Offline collection upon new
				command.
				Offline surface scan supported.
				Self-test supported.
				Conveyance Self-test supported.
				Selective Self-test supported.
SMART capabilities:            (0x0003)	Saves SMART data before entering
				power-saving mode.
				Supports SMART auto save timer.
Error logging capability:        (0x01)	Error logging supported.
				General Purpose Logging supported.
Short self-test routine 
recommended polling time: 	 (   2) minutes.
Extended self-test routine
recommended polling time: 	 (  91) minutes.
Conveyance self-test routine
recommended polling time: 	 (   5) minutes.
SCT capabilities: 	       (0x303f)	SCT Status supported.
				SCT Error Recovery Control supported.
				SCT Feature Control supported.
				SCT Data Table supported.

SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x002f   200   200   051    Pre-fail  Always       -       0
  3 Spin_Up_Time            0x0027   177   173   021    Pre-fail  Always       -       2108
  4 Start_Stop_Count        0x0032   096   096   000    Old_age   Always       -       4467
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x002e   100   253   000    Old_age   Always       -       0
  9 Power_On_Hours          0x0032   066   066   000    Old_age   Always       -       24829
10 Spin_Retry_Count        0x0033   100   100   051    Pre-fail  Always       -       0
11 Calibration_Retry_Count 0x0032   100   100   000    Old_age   Always       -       0
12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       1601
192 Power-Off_Retract_Count 0x0032   200   200   000    Old_age   Always       -       423
193 Load_Cycle_Count        0x0032   157   157   000    Old_age   Always       -       131020
194 Temperature_Celsius     0x0022   122   093   000    Old_age   Always       -       25
196 Reallocated_Event_Count 0x0032   200   200   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0032   200   200   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0030   100   253   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       0
200 Multi_Zone_Error_Rate   0x0009   100   253   051    Pre-fail  Offline      -       0

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline       Completed without error       00%     24829         -

SMART Selective self-test log data structure revision number 1
SPAN  MIN_LBA  MAX_LBA  CURRENT_TEST_STATUS
    1        0        0  Not_testing
    2        0        0  Not_testing
    3        0        0  Not_testing
    4        0        0  Not_testing
    5        0        0  Not_testing
Selective self-test flags (0x0):
  After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.

Smart report does indeed look fine.

 

As for plugins, I see you are running torrents. Torrents can cause quite a bit of drive thrashing especially when downloading and/or seeding multiple items.

 

Stop all of your torrent plugins and retest throughput to the cache drive with LAN Speed Test.

52 MB/s Write and 100 MB/s Read with LAN Speed Test

 

80 MB/s to copy a file from the server to my desktop

 

52 MB/s to copy a file from my desktop to the server with the cache drive enabled

 

Are those results pretty normal for a unRAID server with a spinning cache drive?

A bit on the slow side but not bad, especially considering the older drives being used.

 

The speed you can write to the cache drive will be highly dependent on how many torrents you are downloading/seeding, especially if you have a fairly fast internet connection.

So then my understanding is that if I get a SSD drive as a cache drive it should drastically improve the write speeds even with the plugins. Is that correct?

So then my understanding is that if I get a SSD drive as a cache drive it should drastically improve the write speeds even with the plugins. Is that correct?

 

Yes.

Also. our servers seem pretty similar and we have similar plugins (minus SAB). What sort of speeds are you seeing on yours?

Our servers are indeed fairly similar with the exception of the drives. All of my array drives are 1TB/platter 7200 RPM drives, and the cache is a 320GB/platter 7200 RPM drive.  All 3.5" drives.

 

That said, I get the following speeds:

 

Write to protected array - 40MB/s

Read from array - 120MB/s

Write to cache - 85MB/s

Read from cache - 90MB/s

 

Writes to the array is self explanatory, parity slows it down. Reads from the array go as fast as the network can handle, Gb Ethernet is the bottleneck. The bottleneck with the cache is the drive itself, that's pretty much its max read/write speed.