Swap (and low memory) performance

[oshunluvr]

I also think the new generation of SSD's are more stable. My very first SSD (a 32GB cheapo) died after six weeks. The warranty replacement is still running solid after two+ years. Assuming you buy a name brand with good reviews and support (like regular firmware updates) I think you can treat them as you would any hard drive...

...except you won't be waiting for it much!

[vinnywright]

IDK ...my box routinely uses swap to no ill effect ,,,,even right now doing nothing all day (just waiting for me to get home)

vinny@Vinnys-HP-G62:~$ free
total used free shared buffers cached
Mem: 3844396 2997080 847316 0 211656 1776436
-/+ buffers/cache: 1008988 2835408
Swap: 4096536 21364 4075172

I see this ,but the swap use was probable from watching video last night .

I can be recoding video(making DVD's) browsing the web,listening to music or vid's and DLing torrent's with swap being used and see no lag or bad performance hits.

all with 4Gig's RAM and 4Gig's swap (the swap NEVER gets all used) BUT the RAM will get used up almost all the way but some is always kept free.

VINNY

[claydoh]

Hmm I lost a fairly long post here , for me anyway. I'd blame it on tapatalk but I know better

I have a Dell D630, dual core 2gb ram. Sometime in the past year or so, I have had a serious slowdown when the system goes to swap, seemingly before it needs to, and very often. I chalked it up to not enough ram, nepomuk/soprano, chrome, flash, an aged had drive, all the usual suspects. Try running the Gimp with Kontact and a few chrome tabs open and the countdown to uselessness is usually short. Turning stuff off, closing programs down, changing swappiness, etc only prolongs the inevitable if doing something Gimpy or tabby. I just can't pinpoint exactly when or which version I noticed this, but it was not originally like this.10.10 was the earliest release this machine would have seen.

Now my dell is waiting for some surgery, I am using the 2003 Compaq running Lubuntu (lxde) 12.04. I definitely has many fewer services running, however it has only 1gig ram, is a single core, yet I can run chromium and gimp without hitting swap so hard, fast, and furious as I near the limits of available memory.

[SecretCode]

Thanks guys - I'm concluding that it's at least possible for swap to be used without a killer impact on performance. Time for a fresh install ...

[whatthefunk]

Before you do that fresh install (or during) you should make swap bigger if you can.

[vinnywright]

IDK .......but hears a screen capture of my box doing a lot of stuff at once with some swap being used and no lagging ......well hear we go O the sound didn't come through to well but I was tired from splitting loges for this winters fire wood and didn't feel like fixing it

the shot of the desktop right after the cube shots is a multimedia activity with a vid playing in rosa media player

VINNY

[claydoh]

When I get my Dell repaired I will take a screen cap of how useless it is when the drive goes wild using swap. I wonder if it is hardware specific?

Sent from my SCH-I510 using Tapatalk 2

[SteveRiley]

When I get my Dell repaired I will take a screen cap of how useless it is when the drive goes wild using swap. I wonder if it is hardware specific?

Possibly just a poky slow drive or low-bandwidth interface.

My X1, with a 128 GiB SSD:

steve@x1:~$ [B]sudo hdparm -I /dev/sda[/B]

/dev/sda:

ATA device, with non-removable media
        Model Number:       TOSHIBA THNSNC128GCSJ                   
        Serial Number:      81LS11MLTMAZ
        Firmware Revision:  CJLA0201
        Transport:          Serial, ATA8-AST, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6

steve@x1:~$ [B]sudo hdparm -t /dev/sda[/B]

/dev/sda:
 Timing buffered disk reads: 570 MB in  3.00 seconds = 189.96 MB/sec

Would be interesting to see the results of various other configs of members on the forum here. Then, Claydoh, when you get your laptop back, let's see yours.

[SecretCode]

$ sudo hdparm -t /dev/sda

/dev/sda:
 Timing buffered disk reads: 228 MB in  3.00 seconds =  75.98 MB/sec

I never thought this was a fast disk. But your SSD is only 2.5 times as fast?

hdparm -i returns a lot more information than yours - did you quote only some of it?

$ sudo hdparm -I /dev/sda

/dev/sda:

ATA device, with non-removable media
        Model Number:       ST9500325AS                             
        Serial Number:      6VE4CHJJ
        Firmware Revision:  0002SDM1
        Transport:          Serial
Standards:
        Used: unknown (minor revision code 0x0029) 
        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  = 8192 KBytes
        Nominal Media Rotation Rate: 5400
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 = 16
        Advanced power management level: 254
        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
           *    Advanced Power Management feature set
                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
           *    IDLE_IMMEDIATE with UNLOAD
                Write-Read-Verify feature set
           *    WRITE_UNCORRECTABLE_EXT command
           *    {READ,WRITE}_DMA_EXT_GPL commands
           *    Segmented DOWNLOAD_MICROCODE
           *    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
           *    SMART Command Transport (SCT) feature set
           *    SCT Long Sector Access (AC1)
           *    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
        not     frozen
        not     expired: security count
                supported: enhanced erase
        132min for SECURITY ERASE UNIT. 132min for ENHANCED SECURITY ERASE UNIT.
Logical Unit WWN Device Identifier: 5000c5001ec68fb1
        NAA             : 5
        IEEE OUI        : 000c50
        Unique ID       : 01ec68fb1
Checksum: correct

[SteveRiley]

I never thought this was a fast disk. But your SSD is only 2.5 times as fast?

I think the SSD in my X1 isn't necessarily soooooooper fast, and I also suspect the controller is slow. Here's an alternate pass, this time using raw I/O to bypass the page cache (with the --direct parameter):

steve@x1:~$ [B]sudo hdparm -t --direct /dev/sda[/B]

/dev/sda:
 Timing O_DIRECT disk reads: 586 MB in  3.00 seconds = 195.24 MB/sec

Multiple passes of the test show variations anywhere from 188 to 203 MB/sec.

hdparm -i returns a lot more information than yours - did you quote only some of it?

Yes, I included only the first portion, because what appears interesting is the "Transport" line. I figured this was a way to determine which kind of SATA interface the drive is connected to. But yours is very uninformative, offering only the word "serial" and nothing else. I don't understand that.

[SteveRiley]

For comparison, here's my T520, with a Crucial SSD.

First, disk information:

steve@t520:~$ [B]sudo hdparm -I /dev/sda[/B]

/dev/sda:

ATA device, with non-removable media
        Model Number:       M4-CT256M4SSD2
        Serial Number:      000000001140031E608A
        Firmware Revision:  0009
        Transport:          Serial, ATA8-AST, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0

Note the SATA Rev. 3.0, so this is a faster controller.

Now, read timing tests:

steve@t520:~$ [B]sudo hdparm -t /dev/sda[/B]

/dev/sda:
 Timing buffered disk reads: 1142 MB in  3.00 seconds = 380.09 MB/sec


steve@t520:~$ [B]sudo hdparm -t --direct /dev/sda[/B]

/dev/sda:
 Timing O_DIRECT disk reads: 1266 MB in  3.00 seconds = 421.37 MB/sec

Curious observation: when I run the command without --direct, I can hear a very faint "squeaking" sound while the test runs. But with the parameter, there's no noise at all. Since raw I/O bypasses RAM completely, does this mean that my RAM is squeaky? For the sake of comparison, I checked my X1, and it exhibits a similar phenomenon. It is very faint, but noticeable.

[HalationEffect]

I'm running a 128GB SATA3 SSD, but it's connected to my motherboard's SATA2 interface. Here are my results:

sudo hdparm -I /dev/sda

/dev/sda:

ATA device, with non-removable media
        Model Number:       OCZ-AGILITY3                            
        Serial Number:      OCZ-YL89X9LB1I5N0570
        Firmware Revision:  2.22    
        Transport:          Serial, ATA8-AST, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0

sudo hdparm -t /dev/sda

/dev/sda:
 Timing buffered disk reads: 794 MB in  3.01 seconds = 264.02 MB/sec

sudo hdparm -t --direct /dev/sda

/dev/sda:
 Timing O_DIRECT disk reads: 794 MB in  3.00 seconds = 264.45 MB/sec

[vinnywright]

OK I'll join in

sudo hdparm -I /dev/sda
[sudo] password for vinny:

/dev/sda:

ATA device, with non-removable media
Model Number: WDC WD5000BEVT-60A0RT0
Serial Number: WD-WXR1EA0YSR08
Firmware Revision: 02.01A02
Transport: Serial, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6

and

vinny@Vinnys-HP-G62:~$ sudo hdparm -t /dev/sda

/dev/sda:
Timing buffered disk reads: 226 MB in 3.01 seconds = 75.04 MB/sec

and

vinny@Vinnys-HP-G62:~$ sudo hdparm -t --direct /dev/sda

/dev/sda:
Timing O_DIRECT disk reads: 226 MB in 3.00 seconds = 75.33 MB/sec

one day if the SSD fairy comes to my house maby I to can sport a SSD .............LOL

VINNY

[SecretCode]

So why is my "Transport:" information deficient compared to everyone else's?

[claydoh]

My ata laptop drive in my Compaq Evo N610c for comparo

/dev/sda:

ATA device, with non-removable media
Model Number: WDC WD800BEVE-00UYT0
Serial Number: WD-WXH308413191
Firmware Revision: 01.04A01

jake@jake-the-laptop-of-DOOOOOM-muahahahah:~$ sudo hdparm -t /dev/sda

/dev/sda:
Timing buffered disk reads: 140 MB in 3.01 seconds = 46.53 MB/sec

I have NOT had a swap hit like I did on the Dell as of yet, though I am going somewhat easy on the old guy. I will see what the used motherboard brings me in a few days.