Jump to content
abarbarian

Disk Speed Test (Read/Write): SSD Performance in Linux

Recommended Posts

abarbarian

Whenever I want to buy a new piece of kit I always look for benchmarking test results comparing several different manufacturers offerings. Once I have the kit I run a few tests on my own pc to see if they match the test results posted in articles.

Sometimes you do not need to buy the latest and best kit to get a decent up grade. Sometimes buying a middle of the road piece of kit will save you a load of loot and you will still get almost the same performance as a high end high cost piece of kit.

 

So here is a simple way to benchmark a ssd which will only take a few moments of your time. If you run the tests and post the output here, along with the operating system used and the ssd model and manufacturer information it may help someone to save some loot. At the very least it would be a fun geeky thing to do. An we are all dye in the wool geeks ain't we.

 

https://www.shellhacks.com/disk-speed-test-read-write-hdd-ssd-perfomance-linux/

 

Samsung SP951 NVME 256 GB with Arch 64 on 20/100 GB root/home partitions.

 

$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.436009 s, 2.5 GB/s

 

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.0811887 s, 13.2 GB/s

 

$ sudo /sbin/sysctl -w vm.drop_caches=3

vm.drop_caches = 3

 

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.559377 s, 1.9 GB/s

 

$ sudo hdparm -Tt /dev/nvme0n1p5

/dev/nvme0n1p5:

Timing cached reads: 34302 MB in 1.99 seconds = 17217.89 MB/sec

Timing buffered disk reads: 4330 MB in 3.00 seconds = 1443.31 MB/sec

 

Results can vary so it is best to run the tests several times and cherry pick the top results :blissysmile:

 

I'll try and post some results from a different ssd later on. :breakfast:

Share this post


Link to post
Share on other sites
V.T. Eric Layton

SSD. Ah... one day. Currently, above my pay grade. I'm stuck with mechanical relics that whir, hum, and occasionally go clank, boing, and bang.

Share this post


Link to post
Share on other sites
abarbarian

mechanical relics that whir, hum, and occasionally go clank, boing, and bang.

 

 

:hysterical: :hysterical: :hysterical: :Laughing: :Laughing: :Laughing:

  • Like 1

Share this post


Link to post
Share on other sites
raymac46

Haven't done any tests yet on my creaky 2013 model Toshiba SSD but I suspect you have already won the sweepstakes with that NVME PCI connected drive.

In my view, the big advantage of even a SATA connected SSD is in the performance boost you get over the traditional HDD. Conventional SATA III SSD models offer great value nowadays.

My older hardware does not support the M.2 variety of drive but most of it has a SATA III connection - and I have made use of it. Both desktops, a Thinkpad and an old Toshiba netbook rock some sort of SSD. In fact, the only thing that makes the netbook usable is the presence of a cheap SSD.

Share this post


Link to post
Share on other sites
raymac46

OK Here goes:

Toshiba THNSNH128GBST Linux Mint 19 64 Bit

 

ray@ray-basement-SSD ~ $ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync1024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.942638 s, 1.1 GB/sray@ray-basement-SSD ~ $ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.273185 s, 3.9 GB/sray@ray-basement-SSD ~ $ sudo /sbin/sysctl -w vm.drop_caches=3[sudo] password for ray:vm.drop_caches = 3ray@ray-basement-SSD ~ $ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.96241 s, 547 MB/sray@ray-basement-SSD ~ $ sudo hdparm -Tt /dev/sda/dev/sda:Timing cached reads:   5258 MB in  2.00 seconds = 2629.81 MB/secTiming buffered disk reads: 1570 MB in  3.00 seconds = 522.69 MB/sec

 

Share this post


Link to post
Share on other sites
abarbarian

OK Here goes:

Toshiba THNSNH128GBST Linux Mint 19 64 Bit

 

ray@ray-basement-SSD ~ $ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync1024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.942638 s, 1.1 GB/sray@ray-basement-SSD ~ $ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.273185 s, 3.9 GB/sray@ray-basement-SSD ~ $ sudo /sbin/sysctl -w vm.drop_caches=3[sudo] password for ray:vm.drop_caches = 3ray@ray-basement-SSD ~ $ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.96241 s, 547 MB/sray@ray-basement-SSD ~ $ sudo hdparm -Tt /dev/sda/dev/sda:Timing cached reads: 5258 MB in 2.00 seconds = 2629.81 MB/secTiming buffered disk reads: 1570 MB in 3.00 seconds = 522.69 MB/sec

 

 

 

Considering 128 GB ssd's always run slower than the larger models those are darn good results.

 

In real life usage I only notice a very small difference in speed between an ordinary ssd and my spiffy nvme drive. The one area it does make a difference is in games. Loading games is so darn quick and very noticeable in the slightly older games. It can be quite disconcerting at times as I have been so used to taking a breather after dying that instant reincarnation comes as quite a shock. :laugh:

Share this post


Link to post
Share on other sites
raymac46

One more result from my Thinkpad T430.

Samsung MZ7PC128HAFU-00H1 Debian Stretch 64 Bit.

 

 

ray@ray-debian-T430:~$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync1024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.33347 s, 805 MB/sray@ray-debian-T430:~$ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.185669 s, 5.8 GB/sray@ray-debian-T430:~$ sudo /sbin/sysctl -w vm.drop_caches=3[sudo] password for ray:vm.drop_caches = 3ray@ray-debian-T430:~$ dd if=tempfile of=/dev/null bs=1M count=10241024+0 records in1024+0 records out1073741824 bytes (1.1 GB, 1.0 GiB) copied, 2.22186 s, 483 MB/sray@ray-debian-T430:~$ sudo hdparm -Tt /dev/sda/dev/sda:Timing cached reads:   12662 MB in  2.00 seconds = 6343.71 MB/secTiming buffered disk reads: 1404 MB in  3.00 seconds = 467.70 MB/sec

 

 

http://ssd.userbenchmark.com/SpeedTest/3368/SAMSUNG-MZ7PC128HAFU-000H1

  • Like 1

Share this post


Link to post
Share on other sites
raymac46

My only games are train sims I run in Windows but I have them on an SSD and they do load up quicker. These are rather massive sets of assets and scenes and I don't worry about loading speed TOO much since it takes more than an hour to drive a Castle class steam loco from Oxford to London Paddington. It is never a good idea to die in a train sim. :medic:

 

castle.jpg

Edited by raymac46
  • Like 1

Share this post


Link to post
Share on other sites
securitybreach

I ran these several times and the results are all over the board. This may have been reliable with spinning drives but the results can vary so much that you cannot even count on these tests anymore. I had some vary from 14 GB/s to 800 MB/s.

Share this post


Link to post
Share on other sites
raymac46
I had some vary from 14 GB/s to 800 MB/s.

Interesting. I assume you have one of those NVME drives. Mine are SATA III and I only ran one test. Results seem plausible if a bit optimistic.

Share this post


Link to post
Share on other sites
securitybreach
I had some vary from 14 GB/s to 800 MB/s.

Interesting. I assume you have one of those NVME drives. Mine are SATA III and I only ran one test. Results seem plausible if a bit optimistic.

 

Nah, its Samsung EVO 860 250gb but I also have 9 drives in this machine so it could be looking at various drive ??

Share this post


Link to post
Share on other sites
abarbarian
$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.7818 s, 603 MB/s

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.110958 s, 9.7 GB/s

$ sudo /sbin/sysctl -w vm.drop_caches=3

vm.drop_caches = 3

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.563008 s, 1.9 GB/s

$ sudo hdparm -Tt /dev/nvme0n1p5

 

/dev/nvme0n1p5:

Timing cached reads: 26446 MB in 1.99 seconds = 13270.18 MB/sec

Timing buffered disk reads: 3494 MB in 3.00 seconds = 1162.84 MB/sec

$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.44164 s, 745 MB/s

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.108811 s, 9.9 GB/s

$ sudo /sbin/sysctl -w vm.drop_caches=3

vm.drop_caches = 3

$ dd if=tempfile of=/dev/null bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.606591 s, 1.8 GB/s

$ sudo hdparm -Tt /dev/nvme0n1p5

 

/dev/nvme0n1p5:

Timing cached reads: 25598 MB in 1.99 seconds = 12843.23 MB/sec

Timing buffered disk reads: 4162 MB in 3.00 seconds = 1387.26 MB/sec

 

I ran two tests in quick succession and got pretty consistent results. They are down on my first set as I left BOINC running this time which will have had an impact albeit a small one.

 

These tests run on hdd's aswell so if any one is running their os on a hdd it would be nice to see results from one of those. It might take a tad longer to run the tests but the results would be interesting to see. :devil:

Share this post


Link to post
Share on other sites
securitybreach

I ran two tests in quick succession and got pretty consistent results. They are down on my first set as I left BOINC running this time which will have had an impact albeit a small one.

 

These tests run on hdd's aswell so if any one is running their os on a hdd it would be nice to see results from one of those. It might take a tad longer to run the tests but the results would be interesting to see. :devil:

 

I dunno then :ermm:

Share this post


Link to post
Share on other sites
V.T. Eric Layton

SSD. Ah... one day. Currently, above my pay grade. I'm stuck with mechanical relics that whir, hum, and occasionally go clank, boing, and bang.

 

Hmm... my Secret (early) Santa dropped ship something to me from the North Pole ...er, I mean NewEgg today. I can no longer say that I do not have an SSD. The Ol' Luddite moves into the new millennium. ;)

 

Oh, and it's actually compatible with my ericsbane07 GigaByte mobo. When my next system overhaul comes up, I'll be installing this in that system. YAY!

 

Thank you, Secret Santa! I never met so many wonderful folks till I got on the Internets. :)

 

===

 

And just for fun, I found some specs for my old clunky mechanical drives. They're respectable specs... for old clunky mechanical drives, of course; not for an SSD.

 

ericsbane07 hard drive specs:

 

WD Blue 500G SATA III (Slackware64 14.1)

Maxtor 500G SATA II (Slackware64 14.1 rsync backup)

WD 320G SATA II (MS Windows 7 Enterprise SP2 + common storage)

WD 320G SATA II (MS Windows 7 Enterprise SP2 + common storage - Clonezilla mirror)

 

ericsbane07 hard drive performance:

 

---Drive Tests---

 

Key: sda = Western Digital 500G/7200 primary drive with Slackware OS + storage

sdb = Maxtor 500G/7200 backup drive with sync'd Slackware OS + storage (rsync)

sdc = Western Digital 320G/7200 secondary drive with MS Win 7 + common_storage

sdd = Western Digital 320G/7200 backup drive with MS Win 7 image + common_storage image (Clonezilla)

 

Transfer rates:

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sda

 

/dev/sda:

Timing cached reads: 7218 MB in 2.00 seconds = 3609.82 MB/sec

Timing buffered disk reads: 362 MB in 3.01 seconds = 120.30 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdb

 

/dev/sdb:

Timing cached reads: 7176 MB in 2.00 seconds = 3589.18 MB/sec

Timing buffered disk reads: 232 MB in 3.01 seconds = 77.04 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdc

 

/dev/sdc:

Timing cached reads: 7180 MB in 2.00 seconds = 3591.10 MB/sec

Timing buffered disk reads: 192 MB in 3.03 seconds = 63.38 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdd

 

/dev/sdd:

Timing cached reads: 7252 MB in 2.00 seconds = 3626.28 MB/sec

Timing buffered disk reads: 312 MB in 3.01 seconds = 103.82 MB/sec

 

SATA Capabilities:

 

root@ericsbane07/home/vtel57:# hdparm -I /dev/sda | grep -i speed

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

* Gen3 signaling speed (6.0Gb/s)

root@ericsbane07/home/vtel57:# hdparm -I /dev/sdb | grep -i speed

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

root@ericsbane07/home/vtel57:# hdparm -I /dev/sdc | grep -i speed

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

root@ericsbane07/home/vtel57:# hdparm -I /dev/sdd | grep -i speed

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

 

Drive Information:

 

/dev/sda:

 

ATA device, with non-removable media

Model Number: WDC WD5000AAKX-22ERMA0

Serial Number: WD-WCC2EMJ12973

Firmware Revision: 17.01H17

Transport: Serial, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0

Standards:

Supported: 8 7 6 5

Likely used: 8

Configuration:

Logical max current

cylinders 16383 16383

heads 16 16

sectors/track 63 63

--

CHS current addressable sectors: 16514064

LBA user addressable sectors: 268435455

LBA48 user addressable sectors: 976773168

Logical/Physical Sector size: 512 bytes

device size with M = 1024*1024: 476940 MBytes

device size with M = 1000*1000: 500107 MBytes (500 GB)

cache/buffer size = 16384 KBytes

Capabilities:

LBA, IORDY(can be disabled)

Queue depth: 32

Standby timer values: spec'd by Standard, with device specific minimum

R/W multiple sector transfer: Max = 16 Current = 0

DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6

Cycle time: min=120ns recommended=120ns

PIO: pio0 pio1 pio2 pio3 pio4

Cycle time: no flow control=120ns IORDY flow control=120ns

Commands/features:

Enabled Supported:

* SMART feature set

Security Mode feature set

* Power Management feature set

* Write cache

* Look-ahead

* Host Protected Area feature set

* WRITE_BUFFER command

* READ_BUFFER command

* NOP cmd

* DOWNLOAD_MICROCODE

Power-Up In Standby feature set

* SET_FEATURES required to spinup after power up

SET_MAX security extension

* 48-bit Address feature set

* Device Configuration Overlay feature set

* Mandatory FLUSH_CACHE

* FLUSH_CACHE_EXT

* SMART error logging

* SMART self-test

* General Purpose Logging feature set

* 64-bit World wide name

* {READ,WRITE}_DMA_EXT_GPL commands

* Segmented DOWNLOAD_MICROCODE

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

* Gen3 signaling speed (6.0Gb/s)

* Native Command Queueing (NCQ)

* Host-initiated interface power management

* Phy event counters

* NCQ priority information

* DMA Setup Auto-Activate optimization

* Software settings preservation

* SMART Command Transport (SCT) feature set

* SCT Read/Write Long (AC1), obsolete

* SCT Write Same (AC2)

* SCT Features Control (AC4)

* SCT Data Tables (AC5)

unknown 206[12] (vendor specific)

unknown 206[13] (vendor specific)

Security:

Master password revision code = 65534

supported

not enabled

not locked

frozen

not expired: security count

supported: enhanced erase

88min for SECURITY ERASE UNIT. 88min for ENHANCED SECURITY ERASE UNIT.

Logical Unit WWN Device Identifier: 50014ee208f32a87

NAA : 5

IEEE OUI : 0014ee

Unique ID : 208f32a87

Checksum: correct

 

===

 

/dev/sdb:

 

ATA device, with non-removable media

Model Number: MAXTOR STM3500630AS

Serial Number: 6QG0YEY5

Firmware Revision: 3.AAE

Standards:

Supported: 7 6 5 4

Likely used: 7

Configuration:

Logical max current

cylinders 16383 16383

heads 16 16

sectors/track 63 63

--

CHS current addressable sectors: 16514064

LBA user addressable sectors: 268435455

LBA48 user addressable sectors: 976773168

Logical Sector size: 512 bytes

Physical Sector size: 512 bytes

device size with M = 1024*1024: 476940 MBytes

device size with M = 1000*1000: 500107 MBytes (500 GB)

cache/buffer size = 16384 KBytes

Capabilities:

LBA, IORDY(can be disabled)

Queue depth: 32

Standby timer values: spec'd by Standard, no device specific minimum

R/W multiple sector transfer: Max = 16 Current = ?

Recommended acoustic management value: 254, current value: 0

DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6

Cycle time: min=120ns recommended=120ns

PIO: pio0 pio1 pio2 pio3 pio4

Cycle time: no flow control=120ns IORDY flow control=120ns

Commands/features:

Enabled Supported:

* SMART feature set

Security Mode feature set

* Power Management feature set

* Write cache

* Look-ahead

* Host Protected Area feature set

* WRITE_BUFFER command

* READ_BUFFER command

* DOWNLOAD_MICROCODE

SET_MAX security extension

* 48-bit Address feature set

* Device Configuration Overlay feature set

* Mandatory FLUSH_CACHE

* FLUSH_CACHE_EXT

* SMART error logging

* SMART self-test

* General Purpose Logging feature set

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

* Native Command Queueing (NCQ)

* Phy event counters

Device-initiated interface power management

* Software settings preservation

Security:

Master password revision code = 65534

supported

not enabled

not locked

frozen

not expired: security count

not supported: enhanced erase

Checksum: correct

 

===

 

/dev/sdc:

 

ATA device, with non-removable media

Model Number: WDC WD3200KS-75PFB0

Serial Number: WD-WCAPD4388243

Firmware Revision: 21.00M21

Standards:

Supported: 7 6 5 4

Likely used: 8

Configuration:

Logical max current

cylinders 16383 16383

heads 16 16

sectors/track 63 63

--

CHS current addressable sectors: 16514064

LBA user addressable sectors: 268435455

LBA48 user addressable sectors: 625142448

Logical/Physical Sector size: 512 bytes

device size with M = 1024*1024: 305245 MBytes

device size with M = 1000*1000: 320072 MBytes (320 GB)

cache/buffer size = 16384 KBytes

Capabilities:

LBA, IORDY(can be disabled)

Queue depth: 32

Standby timer values: spec'd by Standard, with device specific minimum

R/W multiple sector transfer: Max = 16 Current = 0

Recommended acoustic management value: 128, current value: 128

DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6

Cycle time: min=120ns recommended=120ns

PIO: pio0 pio1 pio2 pio3 pio4

Cycle time: no flow control=120ns IORDY flow control=120ns

Commands/features:

Enabled Supported:

* SMART feature set

Security Mode feature set

* Power Management feature set

* Write cache

* Look-ahead

* Host Protected Area feature set

* WRITE_BUFFER command

* READ_BUFFER command

* NOP cmd

* DOWNLOAD_MICROCODE

Power-Up In Standby feature set

* SET_FEATURES required to spinup after power up

SET_MAX security extension

Automatic Acoustic Management feature set

* 48-bit Address feature set

* Device Configuration Overlay feature set

* Mandatory FLUSH_CACHE

* FLUSH_CACHE_EXT

* SMART error logging

* SMART self-test

* General Purpose Logging feature set

* WRITE_{DMA|MULTIPLE}_FUA_EXT

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

* Native Command Queueing (NCQ)

* Host-initiated interface power management

* Phy event counters

* DMA Setup Auto-Activate optimization

Device-initiated interface power management

* Software settings preservation

* SMART Command Transport (SCT) feature set

* SCT Read/Write Long (AC1), obsolete

* SCT Write Same (AC2)

* SCT Error Recovery Control (AC3)

* SCT Features Control (AC4)

* SCT Data Tables (AC5)

unknown 206[12] (vendor specific)

Security:

Master password revision code = 65534

supported

not enabled

not locked

frozen

not expired: security count

not supported: enhanced erase

Checksum: correct

 

===

 

/dev/sdd:

 

ATA device, with non-removable media

Model Number: WDC WD3200AAKS-61L9A0

Serial Number: WD-WCAV2K446493

Firmware Revision: 01.03E01

Transport: Serial, SATA 1.0a, SATA II Extensions, SATA Rev 2.5

Standards:

Supported: 8 7 6 5

Likely used: 8

Configuration:

Logical max current

cylinders 16383 16383

heads 16 16

sectors/track 63 63

--

CHS current addressable sectors: 16514064

LBA user addressable sectors: 268435455

LBA48 user addressable sectors: 625142448

Logical/Physical Sector size: 512 bytes

device size with M = 1024*1024: 305245 MBytes

device size with M = 1000*1000: 320072 MBytes (320 GB)

cache/buffer size = 16384 KBytes

Capabilities:

LBA, IORDY(can be disabled)

Queue depth: 32

Standby timer values: spec'd by Standard, with device specific minimum

R/W multiple sector transfer: Max = 16 Current = 0

Recommended acoustic management value: 128, current value: 128

DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6

Cycle time: min=120ns recommended=120ns

PIO: pio0 pio1 pio2 pio3 pio4

Cycle time: no flow control=120ns IORDY flow control=120ns

Commands/features:

Enabled Supported:

* SMART feature set

Security Mode feature set

* Power Management feature set

* Write cache

* Look-ahead

* Host Protected Area feature set

* WRITE_BUFFER command

* READ_BUFFER command

* NOP cmd

* DOWNLOAD_MICROCODE

Power-Up In Standby feature set

* SET_FEATURES required to spinup after power up

SET_MAX security extension

Automatic Acoustic Management feature set

* 48-bit Address feature set

* Device Configuration Overlay feature set

* Mandatory FLUSH_CACHE

* FLUSH_CACHE_EXT

* SMART error logging

* SMART self-test

Media Card Pass-Through

* General Purpose Logging feature set

* 64-bit World wide name

* URG for READ_STREAM[_DMA]_EXT

* URG for WRITE_STREAM[_DMA]_EXT

* Segmented DOWNLOAD_MICROCODE

* Gen1 signaling speed (1.5Gb/s)

* Gen2 signaling speed (3.0Gb/s)

* Native Command Queueing (NCQ)

* Host-initiated interface power management

* Phy event counters

* DMA Setup Auto-Activate optimization

* Software settings preservation

* SMART Command Transport (SCT) feature set

* SCT Read/Write Long (AC1), obsolete

* SCT Write Same (AC2)

* SCT Features Control (AC4)

* SCT Data Tables (AC5)

unknown 206[12] (vendor specific)

unknown 206[13] (vendor specific)

Security:

Master password revision code = 65534

supported

not enabled

not locked

frozen

not expired: security count

supported: enhanced erase

60min for SECURITY ERASE UNIT. 60min for ENHANCED SECURITY ERASE UNIT.

Logical Unit WWN Device Identifier: 50014ee1ad2f4473

NAA : 5

IEEE OUI : 0014ee

Unique ID : 1ad2f4473

Checksum: correct

  • Like 3

Share this post


Link to post
Share on other sites
securitybreach

SSD. Ah... one day. Currently, above my pay grade. I'm stuck with mechanical relics that whir, hum, and occasionally go clank, boing, and bang.

 

Hmm... my Secret (early) Santa dropped ship something to me from the North Pole ...er, I mean NewEgg today. I can no longer say that I do not have an SSD. The Ol' Luddite moves into the new millennium. ;)

 

Oh, and it's actually compatible with my ericsbane07 GigaByte mobo. When my next system overhaul comes up, I'll be installing this in that system. YAY!

 

Thank you, Secret Santa! I never met so many wonderful folks till I got on the Internets. :)

 

Nice :thumbup:

  • Like 2

Share this post


Link to post
Share on other sites
abarbarian

 

 

Hmm... my Secret (early) Santa dropped ship something to me from the North Pole ...er, I mean NewEgg today. I can no longer say that I do not have an SSD. The Ol' Luddite moves into the new millennium. ;)

 

Oh, and it's actually compatible with my ericsbane07 GigaByte mobo. When my next system overhaul comes up, I'll be installing this in that system. YAY!

 

Thank you, Secret Santa! I never met so many wonderful folks till I got on the Internets. :)

 

===

 

And just for fun, I found some specs for my old clunky mechanical drives. They're respectable specs... for old clunky mechanical drives, of course; not for an SSD.

 

 

Transfer rates:

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sda

 

/dev/sda:

Timing cached reads: 7218 MB in 2.00 seconds = 3609.82 MB/sec

Timing buffered disk reads: 362 MB in 3.01 seconds = 120.30 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdb

 

/dev/sdb:

Timing cached reads: 7176 MB in 2.00 seconds = 3589.18 MB/sec

Timing buffered disk reads: 232 MB in 3.01 seconds = 77.04 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdc

 

/dev/sdc:

Timing cached reads: 7180 MB in 2.00 seconds = 3591.10 MB/sec

Timing buffered disk reads: 192 MB in 3.03 seconds = 63.38 MB/sec

 

root@ericsbane07/home/vtel57:# hdparm -tT /dev/sdd

 

/dev/sdd:

Timing cached reads: 7252 MB in 2.00 seconds = 3626.28 MB/sec

Timing buffered disk reads: 312 MB in 3.01 seconds = 103.82 MB/sec

 

 

Yer secret Santa is a tad early ain't he, but I guess you arn't complaining. :yes:

 

Any pc with a SATA interface will run any ssd as even the latest drives are backwards compatible. They will be limited to the SATA spec on the pc though.

 

https://kb.sandisk.com/app/answers/detail/a_id/8142/~/difference-between-sata-i%2C-sata-ii-and-sata-iii

 

To illustrate this I came across this forum post which shows a ssd replacing a slow 5400 hdd running in a SATA 1 capable pc.

 

https://www.overclock.net/forum/355-ssd/1269313-ssd-s-backwards-compatible-sata-i.html

 

As you can see the speed difference is mind blowing.

 

I realise that you probably know a lot of the above but I included it for any folk who are not so clued up on ssd's.

 

It will be interesting to see how your new drives test results compare to your old hdd's. :fishing:

 

I am puzzled by this comment ,

 

"When my next system overhaul comes up, I'll be installing this in that system."

 

You have clones of your os's so I would have thought the question would have been , " I wonder how long it will take to run a clone operation ", as it will only take a couple of minutes at most to plug in the correct cables. :harhar:

  • Like 1

Share this post


Link to post
Share on other sites
raymac46

Right. It isn't a big deal to introduce an SSD. I even managed it with Windows by cloning the drive and then transferring data files to the reformatted HDD. With Linux I backed up the data, reinstalled the O/S on the SSD and then restored the data to the HDD after reformatting it.

My Windows desktop has two SATA III ports and 4 SATA II so I used the SATA III for the SSD and HDD obviously. The Gigabyte mobo on my Linux desktop is all SATA III so no problems.

Edited by raymac46
  • Like 1

Share this post


Link to post
Share on other sites
securitybreach

I am puzzled by this comment ,

 

"When my next system overhaul comes up, I'll be installing this in that system."

 

You have clones of your os's so I would have thought the question would have been , " I wonder how long it will take to run a clone operation ", as it will only take a couple of minutes at most to plug in the correct cables. :harhar:

 

I do not think that he realizes how much of a speed increase that his system will get.

  • Like 1

Share this post


Link to post
Share on other sites
raymac46

Here is an HDD on an old laptop running MX-17 64 bit.

 

ray@mx1-ray:~
$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.68288 s, 638 MB/s
ray@mx1-ray:~
$ dd if=tempfile of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.416644 s, 2.6 GB/s
ray@mx1-ray:~
$ sudo /sbin/sysctl -w vm.drop_caches=3
vm.drop_caches = 3
ray@mx1-ray:~
$ dd if=tempfile of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 36.9473 s, 29.1 MB/s
ray@mx1-ray:~
$ sudo hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads:   3470 MB in  2.00 seconds = 1735.65 MB/sec
Timing buffered disk reads: 322 MB in  3.01 seconds = 107.13 MB/sec

  • Like 1

Share this post


Link to post
Share on other sites
abarbarian

I am puzzled by this comment ,

 

"When my next system overhaul comes up, I'll be installing this in that system."

 

You have clones of your os's so I would have thought the question would have been , " I wonder how long it will take to run a clone operation ", as it will only take a couple of minutes at most to plug in the correct cables. :harhar:

 

I do not think that he realizes how much of a speed increase that his system will get.

 

Maybe he is worried that he will not be able to take forty winks as Stalker reloads when he dies in it so is putting of the upgrade. :Laughing:

 

 

 

Here is an HDD on an old laptop running MX-17 64 bit.

 

ray@mx1-ray:~
$ sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.68288 s, 638 MB/s
ray@mx1-ray:~
$ dd if=tempfile of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.416644 s, 2.6 GB/s
ray@mx1-ray:~
$ sudo /sbin/sysctl -w vm.drop_caches=3
vm.drop_caches = 3
ray@mx1-ray:~
$ dd if=tempfile of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 36.9473 s, 29.1 MB/s
ray@mx1-ray:~
$ sudo hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 3470 MB in 2.00 seconds = 1735.65 MB/sec
Timing buffered disk reads: 322 MB in 3.01 seconds = 107.13 MB/sec

 

Darn fine speeds there mate. I have not been able to reproduce my initial WRITE speeds and in fact at the moment my WRITE speeds are almost three times slower than your hdd writes.

I have a feeling that as my root only has 100 MB of free space this is affecting writes. So maybe a half full cup is better than a full one in this case :thudna5:

 

[longship@15:38:40 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.28897 s, [b]833 MB/s[/b]
[longship@15:38:42 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.2054 s, [b]891 MB/s[/b]
[longship@15:38:45 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.34938 s, [b]796 MB/s[/b]
[longship@15:38:50 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 1.88578 s,[b] 569 MB/s[/b]
[longship@15:38:54 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.34639 s, [b]321 MB/s[/b]
[longship@15:38:58 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.90552 s, [b]275 MB/s[/b]
[longship@15:39:05 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out2
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 4.04011 s, [b]2[/b][b]66 MB/s[/b]
[longship@15:39:11 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.96469 s, [b]271 MB/s[/b]
[longship@15:39:36 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.76244 s, [b]285 MB/s[/b]
[longship@15:39:42 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.84389 s, [b]279 MB/s[/b]
[longship@15:39:47 ~]$  sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 3.84847 s, [b]279 MB/s[/b]

  • Like 1

Share this post


Link to post
Share on other sites
raymac46

I am not doing anything exotic I don't think. All machines have a minimum of 8 GB RAM, the desktop has 16 GB. I don't have a root partition so the whole drive is available. Desktop has rather pedestrian quad-core AMD APU one laptop has an Ivy Bridge i5 and the one with the HDD has a dual-core AMD Trinity APU.

Read speeds seem rather high in all cases and the write speed on the HDD is pretty low compared to the other drives.

Share this post


Link to post
Share on other sites
V.T. Eric Layton

Nah... I'm just lazy and don't feel like pulling the covers off this machine. It's working fine. I'm happy. Who needs fast? I'm old. I like things slow and understandable. ;)

 

I do have a question, though...

 

Can you rsync and/or Clonezilla from an SSD to a mechanical drive?

Share this post


Link to post
Share on other sites
securitybreach

Can you rsync and/or Clonezilla from an SSD to a mechanical drive?

 

Yes, just tweak the ssd when you are done.

  • Like 1

Share this post


Link to post
Share on other sites
V.T. Eric Layton

Ah... many things I have to learn, I s'pose. If I remember correctly, there are already some SSD threads around here that explain a lot of it. :)

Share this post


Link to post
Share on other sites
securitybreach

Ah... many things I have to learn, I s'pose. If I remember correctly, there are already some SSD threads around here that explain a lot of it. :)

 

Just follow this as it covers TRIM and such: https://wiki.archlinux.org/index.php/Solid_state_drive

  • Like 1

Share this post


Link to post
Share on other sites
abarbarian

Ah... many things I have to learn, I s'pose. If I remember correctly, there are already some SSD threads around here that explain a lot of it. :)

 

I had forgotten that you need to do some background reading before you installed and migrated to a ssd. I found the following articles very helpful as starter information when I made the transition as they contain clear and step by step information which was very helpful to a novice user like myself. They are well worth the ten or so minutes they need to read them.

 

http://www.linux-magazine.com/Issues/2015/172/Tuning-Your-SSD

 

http://www.linux-magazine.com/Issues/2014/164/Migrating-to-an-SSD

 

Re-reading the above made me realise that I had missed out on getting clued up on the "lazytime" mount option so I am off to investigate. I may very well not need it as Arch can utilise a systemd unit for trim which runs at default once a week and I have already set this up.

I also have not set up several of the optimisations for ssd's so I'll be investigating them as well. I think I have set up my Windows to run properly on a ssd but I will check that out also just to be sure.

 

Solid state drives make everything run faster and more smoothly, but you can squeeze out even more performance with some practical optimization steps.

 

all of the above links are a tad old but almost all the information is still relevant.

 

I started this thread as a bit of fun however whilst it has been fun it has also led to discovery of more useful information. You can just pop a ssd in your pc and do a basic set up and run with it quite happily or you can scratch that geeky itch and fine tune it to gain every last little bit of advantage from it.

 

I also think I have solved the mystery of why my nvme drive was posting slower write results than other folks ssd's. My drive was almost full which will have some impact. I also leave my pc on 24/7 as I run Boinc for the WCG which results in a load of reads and writes 24/7 this will also have an impact. I have set up the systemd unit to run at the default weekly time so it may be beneficial for me to set this to run more frequently.

 

Happy speeding folks and don't burn out that old grey cell :harhar:

  • Like 2

Share this post


Link to post
Share on other sites
abarbarian

Ah... many things I have to learn, I s'pose. If I remember correctly, there are already some SSD threads around here that explain a lot of it. :)

 

Just follow this as it covers TRIM and such: https://wiki.archlin...lid_state_drive

 

I may be wrong here but I do not think Eric's Slackware uses systemd so he will not be able to use the service.timer for trim. He should be able to set up a cron job with fstrim. It seems this is a much better way than utilising the "discard" option in fstab.

I have just made some changes to my fstab,

 

#/dev/nvme0n1p4 LABEL=ArchRoot
UUID=facb8acb-8d5d-4d91-8630-31a781a95d44    /		     ext4	      rw,relatime,lazytime,data=ordered    0 1

 

I added the "lazytime" option after extensive reading on the subject. My trim is done via systemd. An I think I am going to change from UUID to part-UUID later on after I have had a much needed cup of tea.

 

If anyone spots faults with my fstab entry do let me know. :breakfast:

  • Like 1

Share this post


Link to post
Share on other sites
securitybreach

Looks good to me but I have never heard of or used the "lazytime" option....

Share this post


Link to post
Share on other sites
abarbarian

Looks good to me but I have never heard of or used the "lazytime" option....

 

Never heard of lazytime :w00t: Blimey and I thought you were right at the cutting edge. Guess we will have to change your nickname from Minute Man to Missed It Man :Laughing: As an x hippy I am familiar with all aspects of lazytime :laugh:

  • Like 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...