NASLite Network Attached Storage

ok, so AFP is the clear winner on performance and using diskwriggler I get this.

AFP
aarons-mac-pro:~ aaron$ /Volumes/store/Downloads/diskWriggler-1.0.1.osx_ppc/diskWriggler -2K -t -n 300 -o /Volumes/192.168.1.10
Frame Resolution : 2048 x 1556
Bytes / Pixel : 4.00 (padded) (10 bit, 4:4:4 sampling)
Frame Size : 12,746,752 bytes
File Type : One file per frame (3,824,025,600 bytes).

Write Summary : 300.00 frames in 52.08 secs
: 2.78(min) 5.76(avg) 6.39(max) FPS
: 33.81(min) 70.03(avg) 77.67(max) MB/s

Read Summary : 300.00 frames in 46.29 secs
: 5.68(min) 6.48(avg) 6.67(max) FPS
: 69.06(min) 78.79(avg) 81.11(max) MB/s

NFS
aarons-mac-pro:~ aaron$ /Volumes/store/Downloads/diskWriggler-1.0.1.osx_ppc/diskWriggler -2K -t -n 300 -o /Volumes/192.168.1.10
Frame Resolution : 2048 x 1556
Bytes / Pixel : 4.00 (padded) (10 bit, 4:4:4 sampling)
Frame Size : 12,746,752 bytes
File Type : One file per frame (3,824,025,600 bytes).

Write Summary : 300.00 frames in 64.49 secs
: 2.85(min) 4.65(avg) 6.31(max) FPS
: 34.59(min) 56.55(avg) 76.76(max) MB/s

Read Summary : 300.00 frames in 47.51 secs
: 5.40(min) 6.31(avg) 6.64(max) FPS
: 65.63(min) 76.76(avg) 80.73(max) MB/s

SMB

aarons-mac-pro:~ aaron$ /Volumes/store/Downloads/diskWriggler-1.0.1.osx_ppc/diskWriggler -2K -t -n 300 -o /Volumes/192.168.1.10
Frame Resolution : 2048 x 1556
Bytes / Pixel : 4.00 (padded) (10 bit, 4:4:4 sampling)
Frame Size : 12,746,752 bytes
File Type : One file per frame (3,824,025,600 bytes).

Write Summary : 300.00 frames in 54.10 secs
: 5.07(min) 5.55(avg) 5.86(max) FPS
: 61.65(min) 67.41(avg) 71.28(max) MB/s

Read Summary : 300.00 frames in 50.72 secs
: 5.56(min) 5.92(avg) 6.07(max) FPS
: 67.59(min) 71.91(avg) 73.80(max) MB/s


BUT

How do we solve the slow transfer using finder? Can anyone help?

are you running 10.4.11? are the speeds through the finder any different after a clean reboot?

I use it. It's easy. Don't have to do a thing and it works, great! Speed is not that impressive with the Finder.

But sins I upgraded to version 2, I can't remove some file's?????????????????? Any ideas???????

using leopard but no difference really from tiger. still puzzling why it is so slow. if i connect to my laptop over the exact same network the speeds are amazing. connecting to the nas is appalling. so much so that i barely use it at the moment which is a real shame as i need my raid 5 disks. i'll be testing out some alternatives soon as it appears that this problem still remains.

Then you can discount your network as the problem, I would troubleshoot the hardware configuration of the NASLite box. Your problem is most definitely a hardware issue.

Post your syslog output to see what that reveals.

FYI, I haven't done any benchmarking with Leopard, but the Finder in Tiger and previous versions has issues utilizing extended memory for copying files across the network. This issue was affirmed to me a while back from a Apple engineer.

I haven't had any problems with AFP. Transfer is on par with other protocols so the issue must be with your hardware or network topology. What's your throughput with cifs or ftp? Do you see any TX/RX errors in the network status page?

I have noticed his speed issue with another NAS device I have.
I have a N5200 running V2.00.01 RAID5, 4 drives with 1 hot spare and NasLite V2.06 both have SMB and AFP on (FTP and NFS off)
I also have a W2K server running Software RAID5 three drives. All connect to a unmanaged Netgear Gigabit switch (my managed one died )
I Found via SMB to the NAS servers with OSX 10.3.9 to OSX 10.5.1 speed was always good, constant in both read and write.
I found via AFP diskwriggler also had great read write speeds, but through the finder it was another story. All had great read times, but to write a file had different results depending on the OS. I used the same hardware for this (G5 dual 2.5GHz, 4Gb ram). OSX 10.3.9 was the worst to get a file over via AFP to the NAS servers, 10.4.8 wasn't that much better 10.4.11 improved a lot (still not as good as SMB) 10.5.1 is a little faster again, but again not as good as SMB. It seemed that the copy would start full speed, pause, go, pause, go, pause, etc etc. This happened a lot with 10.3.9-10.4.8. 10.4.11-10.5.2 does pause, but no where near the amount or pause time.
The strange thing is I also tested a FreeNAS for a short time, it also had the same problems.
One thing I found strange was, I never had a problem with the W2K server using AFP with any OSX version. I don't know why MS version of AFP has no speed problems, while Netatalk's AFP (I think that is what FreeNAS, N5200 and NasLite use) had this problem? Anyone?
I am happy with the speeds, NasLite is a little better than both FreeNAS and N5200 for some reason (I think NasLites version has been tweaked?).
OSX to OSX is the fastest network transfers I see, but no surprise there.

I have exactly the same problem, initial transfer starts at around 67mb/s then drops to around 15 then stops (pauses) for a while, then starts the cycle again. I'll see if smb has the same problem.

same here, exactly the same architecture for the network and osx to osx rockets along. i've tried two different network cards in the nas, currently intel MT1000 and still the same exists. If i run it as a windows server, windows-windows or even OSX to windows is much much quicker than NAS. I think I can eliminate the hardware as the bottleneck here, its a raid 5 array which is plenty quick enough, P4 chip (2., 2 gigs of ram. So looking like some kind of software issue.

If we could get NAS2 to run AFP like OSX does? That would be great. My speeds on the network are quicker at a constant 30-32mb/s.

I am using it. I just set it up for my wife's MacBook with osX. I like it a lot better. I have it set so she has read only rights. I have our family's pics on this server, and it keeps down on the oops, I deleted it. Tony thanks for the AFP support. I liked it so much I converted my NASLite-2 USB to a CF install so there is no more USB. I love it. I also bought today a NASLite-2 HDD for my brother and set up a server for him also running off a CF. Keep up the great work.

I bought my USB copy almost a year and a half ago and just never updated it. I went from version 2.02 to 2.06. All the work you did Tony really shows.

Ralph

As requested here is my log. If you can figure it out I'd be one happy customer!

E.g. tonight (using FTP) connects at 45mb/s for 10-20 secs, then drops to 14/15mb/s

Jan 27 18:07:11 user.notice kernel: klogd started: BusyBox v1.01 (2007.03.15-22:03+0000)
Jan 27 18:07:11 user.warn kernel: Linux version 2.4.35.NASLite (root@tzt) (gcc version 3.4.6) #2 Sun Aug 12 09:03:16 EDT 2007
Jan 27 18:07:11 user.info kernel: BIOS-provided physical RAM map:
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 0000000000000000 - 000000000009f800 (usable)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 000000000009f800 - 00000000000a0000 (reserved)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 0000000000100000 - 000000007fff0000 (usable)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 000000007fff0000 - 000000007fff3000 (ACPI NVS)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 000000007fff3000 - 0000000080000000 (ACPI data)
Jan 27 18:07:11 user.warn kernel: BIOS-e820: 00000000fec00000 - 0000000100000000 (reserved)
Jan 27 18:07:11 user.notice kernel: 1151MB HIGHMEM available.
Jan 27 18:07:11 user.notice kernel: 896MB LOWMEM available.
Jan 27 18:07:11 user.warn kernel: On node 0 totalpages: 524272
Jan 27 18:07:11 user.warn kernel: zone(0): 4096 pages.
Jan 27 18:07:11 user.warn kernel: zone(1): 225280 pages.
Jan 27 18:07:11 user.warn kernel: zone(2): 294896 pages.
Jan 27 18:07:11 user.info kernel: ACPI: RSDP (v000 IntelR ) @ 0x000f7ae0
Jan 27 18:07:11 user.info kernel: ACPI: RSDT (v001 IntelR AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x7fff3000
Jan 27 18:07:11 user.info kernel: ACPI: FADT (v001 IntelR AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x7fff3040
Jan 27 18:07:11 user.info kernel: ACPI: MADT (v001 IntelR AWRDACPI 0x42302e31 AWRD 0x00000000) @ 0x7fff7240
Jan 27 18:07:11 user.info kernel: ACPI: DSDT (v001 INTELR AWRDACPI 0x00001000 MSFT 0x0100000e) @ 0x00000000
Jan 27 18:07:11 user.warn kernel: Kernel command line: rw root=/dev/ram0 initrd=NASLite.02 quiet BOOT_IMAGE=NASLite.01
Jan 27 18:07:11 user.info kernel: Initializing CPU#0
Jan 27 18:07:11 user.warn kernel: Detected 2763.485 MHz processor.
Jan 27 18:07:11 user.warn kernel: Console: colour VGA+ 80x25
Jan 27 18:07:11 user.warn kernel: Calibrating delay loop. 5518.13 BogoMIPS
Jan 27 18:07:11 user.info kernel: Memory: 2066420k/2097088k available (2214k kernel code, 30280k reserved, 642k data, 540k init, 1179584k highmem)
Jan 27 18:07:11 user.info kernel: Dentry cache hash table entries: 262144 (order: 9, 2097152 bytes)
Jan 27 18:07:11 user.info kernel: Inode cache hash table entries: 131072 (order: 8, 1048576 bytes)
Jan 27 18:07:11 user.info kernel: Mount cache hash table entries: 512 (order: 0, 4096 bytes)
Jan 27 18:07:11 user.info kernel: Buffer cache hash table entries: 131072 (order: 7, 524288 bytes)
Jan 27 18:07:11 user.warn kernel: Page-cache hash table entries: 524288 (order: 9, 2097152 bytes)
Jan 27 18:07:11 user.info kernel: CPU: Trace cache: 12K uops, L1 D cache: 8K
Jan 27 18:07:11 user.info kernel: CPU: L2 cache: 512K
Jan 27 18:07:11 user.debug kernel: CPU: After generic, caps: bfebfbff 00000000 00000000 00000000
Jan 27 18:07:11 user.debug kernel: CPU: Common caps: bfebfbff 00000000 00000000 00000000
Jan 27 18:07:11 user.warn kernel: CPU: Intel(R) Pentium(R) 4 CPU 2.40GHz stepping 09
Jan 27 18:07:11 user.info kernel: Enabling fast FPU save and restore. done.
Jan 27 18:07:11 user.info kernel: Enabling unmasked SIMD FPU exception support. done.
Jan 27 18:07:11 user.info kernel: Checking 'hlt' instruction. OK.
Jan 27 18:07:11 user.warn kernel: ACPI: IRQ9 SCI: Level Trigger.
Jan 27 18:07:11 user.warn kernel: POSIX conformance testing by UNIFIX
Jan 27 18:07:11 user.warn kernel: mtrr: v1.40 (20010327) Richard Gooch (rgooch@atnf.csiro.au)
Jan 27 18:07:11 user.warn kernel: mtrr: detected mtrr type: Intel
Jan 27 18:07:11 user.info kernel: ACPI: Subsystem revision 20040326
Jan 27 18:07:11 user.info kernel: PCI: PCI BIOS revision 2.10 entry at 0xfba70, last bus=2
Jan 27 18:07:11 user.info kernel: PCI: Using configuration type 1
Jan 27 18:07:11 user.info kernel: ACPI: Interpreter enabled
Jan 27 18:07:11 user.info kernel: ACPI: Using PIC for interrupt routing
Jan 27 18:07:11 user.info kernel: ACPI: System [ACPI] (supports S0 S3 S4 S5)
Jan 27 18:07:11 user.info kernel: ACPI: PCI Root Bridge [PCI0] (00:00)
Jan 27 18:07:11 user.warn kernel: PCI: Probing PCI hardware (bus 00)
Jan 27 18:07:11 user.info kernel: PCI: Ignoring BAR0-3 of IDE controller 00:1f.1
Jan 27 18:07:11 user.warn kernel: Transparent bridge - PCI device 8086:244e
Jan 27 18:07:11 user.debug kernel: ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
Jan 27 18:07:11 user.debug kernel: ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.HUB0._PRT]
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 *5 7 9 10 11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 7 *9 10 11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 7 9 10 *11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 7 9 *10 11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 7 9 *10 11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 7 9 10 11 *12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNK0] (IRQs 3 4 5 7 9 10 *11 12 14 15)
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNK1] (IRQs *3 4 5 7 9 10 11 12 14 15)
Jan 27 18:07:11 user.info kernel: PCI: Probing PCI hardware
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKA] enabled at IRQ 5
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKD] enabled at IRQ 10
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKC] enabled at IRQ 11
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNK1] enabled at IRQ 3
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKB] enabled at IRQ 9
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKE] enabled at IRQ 10
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNKF] enabled at IRQ 12
Jan 27 18:07:11 user.warn kernel: ACPI: PCI Interrupt Link [LNK0] enabled at IRQ 11
Jan 27 18:07:11 user.info kernel: PCI: Using ACPI for IRQ routing
Jan 27 18:07:11 user.info kernel: Linux NET4.0 for Linux 2.4
Jan 27 18:07:11 user.info kernel: Based upon Swansea University Computer Society NET3.039
Jan 27 18:07:11 user.warn kernel: Initializing RT netlink socket
Jan 27 18:07:11 user.warn kernel: Starting kswapd
Jan 27 18:07:11 user.warn kernel: reserved for the highmem bounces
Jan 27 18:07:11 user.info kernel: Journalled Block Device driver loaded
Jan 27 18:07:11 user.info kernel: devfs: v1.12c (20020818) Richard Gooch (rgooch@atnf.csiro.au)
Jan 27 18:07:11 user.info kernel: devfs: boot_options: 0x1
Jan 27 18:07:11 user.info kernel: Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
Jan 27 18:07:11 user.info kernel: ACPI: Power Button (FF) [PWRF]
Jan 27 18:07:11 user.info kernel: ACPI: Processor [CPU0] (supports C1)
Jan 27 18:07:11 user.info kernel: ACPI: Processor [CPU1] (supports C1)
Jan 27 18:07:11 user.info kernel: ACPI: Thermal Zone [THRM] (20 C)
Jan 27 18:07:11 user.warn kernel: pty: 256 Unix98 ptys configured
Jan 27 18:07:11 user.info kernel: Real Time Clock Driver v1.10f
Jan 27 18:07:11 user.info kernel: Floppy drive(s): fd0 is 1.44M
Jan 27 18:07:11 user.warn kernel: keyboard: Timeout - AT keyboard not present?(ed)
Jan 27 18:07:11 user.warn kernel: keyboard: Timeout - AT keyboard not present?(f4)
Jan 27 18:07:11 user.info kernel: FDC 0 is a post-1991 82077
Jan 27 18:07:11 user.warn kernel: RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
Jan 27 18:07:11 user.info kernel: loop: loaded (max 8 devices)
Jan 27 18:07:11 user.info kernel: Intel(R) PRO/1000 Network Driver - version 7.3.20-k4
Jan 27 18:07:11 user.info kernel: Copyright (c) 1999-2006 Intel Corporation.
Jan 27 18:07:11 user.info kernel: e1000: 02:04.0: e1000_probe: (PCI:33MHz:32-bit) 00:07:e9:17:21:e8
Jan 27 18:07:11 user.info kernel: e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection
Jan 27 18:07:11 user.info kernel: e1000: 02:04.1: e1000_probe: (PCI:33MHz:32-bit) 00:07:e9:17:21:e9
Jan 27 18:07:11 user.info kernel: e1000: eth1: e1000_probe: Intel(R) PRO/1000 Network Connection
Jan 27 18:07:11 user.info kernel: dgrs: SW=$Id: dgrs.c,v 1.13 2000/06/06 04:07:00 rick Exp $ FW=Build 550 11/16/96 03:45:15
Jan 27 18:07:11 user.warn kernel: FW Version=$Version$
Jan 27 18:07:11 user.info kernel: pcnet32.c:v1.30h 06.24.2004 tsbogend@alpha.franken.de
Jan 27 18:07:11 user.info kernel: ThunderLAN driver v1.15
Jan 27 18:07:11 user.info kernel: TLAN: 0 devices installed, PCI: 0 EISA: 0
Jan 27 18:07:11 user.info kernel: dmfe: Davicom DM9xxx net driver, version 1.36.4 (2002-01-17)
Jan 27 18:07:11 user.info kernel: ns83820.c: National Semiconductor DP83820 10/100/1000 driver.
Jan 27 18:07:11 user.warn kernel: sk98lin: No adapter found.
Jan 27 18:07:11 user.info kernel: forcedeth.c: Reverse Engineered nForce ethernet driver. Version 0.50.
Jan 27 18:07:11 user.info kernel: Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
Jan 27 18:07:11 user.info kernel: ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
Jan 27 18:07:11 user.info kernel: ICH5: IDE controller at PCI slot 00:1f.1
Jan 27 18:07:11 user.info kernel: ICH5: chipset revision 2
Jan 27 18:07:11 user.info kernel: ICH5: not 100% native mode: will probe irqs later
Jan 27 18:07:11 user.info kernel: ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:DMA
Jan 27 18:07:11 user.info kernel: ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
Jan 27 18:07:11 user.warn kernel: hda: HL-DT-ST GCE-8160B, ATAPI CD/DVD-ROM drive
Jan 27 18:07:11 user.warn kernel: hdb: TSSTcorpCD/DVDW TS-H552B, ATAPI CD/DVD-ROM drive
Jan 27 18:07:11 user.warn kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Jan 27 18:07:11 user.info kernel: SCSI subsystem driver Revision: 1.00
Jan 27 18:07:11 user.info kernel: Loading Adaptec I2O RAID: Version 2.4 Build 5
Jan 27 18:07:11 user.info kernel: Detecting Adaptec I2O RAID controllers.
Jan 27 18:07:11 user.info kernel: Red Hat/Adaptec aacraid driver (1.1-3 Aug 12 2007 09:05:44)
Jan 27 18:07:11 user.warn kernel: scsi: Detection failed (no card)
Jan 27 18:07:11 user.notice kernel: megaraid: v2.10.10.1 (Release Date: Thu Jan 27 16:19:44 EDT 2005)
Jan 27 18:07:11 user.notice kernel: megaraid: found 0x1000:0x1960:bus 2:slot 6:func 0
Jan 27 18:07:11 user.notice kernel: scsi1:Found MegaRAID controller at 0xf884a000, IRQ:11
Jan 27 18:07:11 user.notice kernel: megaraid: [713N:G119] detected 1 logical drives.
Jan 27 18:07:11 user.notice kernel: megaraid: supports extended CDBs.
Jan 27 18:07:11 user.info kernel: megaraid: channel[0] is raid.
Jan 27 18:07:11 user.info kernel: scsi1 : LSI Logic MegaRAID 713N 254 commands 16 targs 4 chans 7 luns
Jan 27 18:07:11 user.notice kernel: scsi1: scanning scsi channel 0 for logical drives.
Jan 27 18:07:11 user.warn kernel: Vendor: MegaRAID Model: LD 0 RAID5 1144G Rev: 713N
Jan 27 18:07:11 user.warn kernel: Type: Direct-Access ANSI SCSI revision: 02
Jan 27 18:07:11 user.notice kernel: scsi1: scanning scsi channel 1 for logical drives.
Jan 27 18:07:11 user.notice kernel: scsi1: scanning scsi channel 2 for logical drives.
Jan 27 18:07:11 user.notice kernel: scsi1: scanning scsi channel 3 for logical drives.
Jan 27 18:07:11 user.notice kernel: scsi1: scanning scsi channel 4 [P0] for physical devices.
Jan 27 18:07:11 user.warn kernel: GDT-HA: Storage RAID Controller Driver. Version: 3.04
Jan 27 18:07:11 user.warn kernel: GDT-HA: Found 0 PCI Storage RAID Controllers
Jan 27 18:07:11 user.warn kernel: 3ware Storage Controller device driver for Linux v1.02.00.037.
Jan 27 18:07:11 user.warn kernel: 3w-xxxx: No cards found.
Jan 27 18:07:11 user.debug kernel: libata version 1.20 loaded.
Jan 27 18:07:11 user.warn kernel: Attached scsi disk sda at scsi1, channel 0, id 0, lun 0
Jan 27 18:07:11 user.warn kernel: SCSI device sda: 2344237056 512-byte hdwr sectors (1200249 MB)
Jan 27 18:07:11 user.info kernel: Partition check:
Jan 27 18:07:11 user.info kernel: /dev/scsi/host1/bus0/target0/lun0: p1
Jan 27 18:07:11 user.info kernel: ohci1394: $Rev: 1045 $ Ben Collins
Jan 27 18:07:11 user.info kernel: ohci1394_0: OHCI-1394 1.1 (PCI): IRQ=[9] MMIO=[e8049000-e80497ff] Max Packet=[2048]
Jan 27 18:07:11 user.info kernel: ohci1394_1: OHCI-1394 1.1 (PCI): IRQ=[10] MMIO=[e8048000-e80487ff] Max Packet=[2048]
Jan 27 18:07:11 user.info kernel: raw1394: /dev/raw1394 device initialized
Jan 27 18:07:11 user.debug kernel: ieee1394: Host added: ID:BUS[0-00:1023] GUID[00508d0000f8900d]
Jan 27 18:07:11 user.info kernel: sbp2: $Rev: 1074 $ Ben Collins
Jan 27 18:07:11 user.err kernel: ieee1394: sbp2: Driver forced to serialize I/O (serialize_io = 1)
Jan 27 18:07:11 user.info kernel: usb.c: registered new driver usbdevfs
Jan 27 18:07:11 user.info kernel: ed new driver hub
Jan 27 18:07:11 user.debug kernel: PCI: Setting latency timer of device 00:1d.7 to 64
Jan 27 18:07:11 user.info kernel: ehci_hcd 00:1d.7: PCI device 8086:24dd
Jan 27 18:07:11 user.info kernel: ehci_hcd 00:1d.7: irq 3, pci mem f8852000
Jan 27 18:07:11 user.info kernel: usb.c: new USB bus registered, assigned bus number 1
Jan 27 18:07:11 user.info kernel: ehci_hcd 00:1d.7: enabled 64bit PCI DMA
Jan 27 18:07:11 user.debug kernel: PCI: cache line size of 128 is not supported by device 00:1d.7
Jan 27 18:07:11 user.info kernel: ehci_hcd 00:1d.7: USB 2.0 enabled, EHCI 1.00, driver 2003-Dec-29/2.4
Jan 27 18:07:11 user.info kernel: hub.c: USB hub found
Jan 27 18:07:11 user.info kernel: hub.c: 8 ports detected
Jan 27 18:07:11 user.info kernel: host/uhci.c: USB Universal Host Controller Interface driver v1.1
Jan 27 18:07:11 user.debug kernel: PCI: Setting latency timer of device 00:1d.0 to 64
Jan 27 18:07:11 user.info kernel: host/uhci.c: USB UHCI at I/O 0xac00, IRQ 5
Jan 27 18:07:11 user.info kernel: usb.c: new USB bus registered, assigned bus number 2
Jan 27 18:07:11 user.info kernel: hub.c: USB hub found
Jan 27 18:07:11 user.info kernel: hub.c: 2 ports detected
Jan 27 18:07:11 user.debug kernel: PCI: Setting latency timer of device 00:1d.1 to 64
Jan 27 18:07:11 user.info kernel: host/uhci.c: USB UHCI at I/O 0xa000, IRQ 10
Jan 27 18:07:11 user.debug kernel: ieee1394: Host added: ID:BUS[1-00:1023] GUID[00023c01010122dd]
Jan 27 18:07:11 user.info kernel: usb.c: new USB bus registered, assigned bus number 3
Jan 27 18:07:11 user.info kernel: hub.c: USB hub found
Jan 27 18:07:11 user.info kernel: hub.c: 2 ports detected
Jan 27 18:07:11 user.debug kernel: PCI: Setting latency timer of device 00:1d.2 to 64
Jan 27 18:07:11 user.info kernel: host/uhci.c: USB UHCI at I/O 0xa400, IRQ 11
Jan 27 18:07:11 user.info kernel: usb.c: new USB bus registered, assigned bus number 4
Jan 27 18:07:11 user.info kernel: hub.c: USB hub found
Jan 27 18:07:11 user.info kernel: hub.c: 2 ports detected
Jan 27 18:07:11 user.debug kernel: PCI: Setting latency timer of device 00:1d.3 to 64
Jan 27 18:07:11 user.info kernel: host/uhci.c: USB UHCI at I/O 0xa800, IRQ 5
Jan 27 18:07:11 user.info kernel: usb.c: new USB bus registered, assigned bus number 5
Jan 27 18:07:11 user.info kernel: hub.c: USB hub found
Jan 27 18:07:11 user.info kernel: hub.c: 2 ports detected
Jan 27 18:07:11 user.info kernel: usb.c: registered new driver usbkbd
Jan 27 18:07:11 user.info kernel: usbkbd.c: :USB HID Boot Protocol keyboard driver
Jan 27 18:07:11 user.info kernel: Initializing USB Mass Storage driver.
Jan 27 18:07:11 user.info kernel: usb.c: registered new driver usb-storage
Jan 27 18:07:11 user.info kernel: USB Mass Storage support registered.
Jan 27 18:07:11 user.info kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Jan 27 18:07:11 user.info kernel: IP Protocols: ICMP, UDP, TCP
Jan 27 18:07:11 user.info kernel: IP: routing cache hash table of 16384 buckets, 128Kbytes
Jan 27 18:07:11 user.info kernel: TCP: Hash tables configured (established 262144 bind 65536)
Jan 27 18:07:11 user.info kernel: NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
Jan 27 18:07:11 user.notice kernel: RAMDISK: NASLite file system found at block 0
Jan 27 18:07:11 user.warn kernel: <6>hub.c: new USB device 00:1d.1-1, assigned address 2
Jan 27 18:07:11 user.warn kernel: <6>hub.c: USB hub found
Jan 27 18:07:11 user.warn kernel: .<6>hub.c: 3 ports detected
Jan 27 18:07:11 user.warn kernel: <6>Freeing initrd memory: 1952k freed
Jan 27 18:07:11 user.warn kernel: VFS: Mounted root (ext2 filesystem).
Jan 27 18:07:11 user.info kernel: Mounted devfs on /dev
Jan 27 18:07:11 user.info kernel: Freeing unused kernel memory: 540k freed
Jan 27 18:07:11 user.info kernel: hub.c: new USB device 00:1d.1-1.3, assigned address 3
Jan 27 18:07:11 user.info kernel: input0: Mitsumi Electric Apple Extended USB Keyboard on usb3:3.0
Jan 27 18:07:11 user.info kernel: e1000: eth1: e1000_watchdog: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX
Jan 27 18:07:11 user.err kernel: ext3: No journal on filesystem on sd(8,1)

It sounds like you're filling up the RAM buffers (quickly) then running at the native throughput of the drives (more slowly), while the buffers are continuously filled / flushed. I also have 2GB (currently, main PC toasted 2 x 1GB DDR2-900.... ) and mine performs in a similar manner.

Did you try SMB? how did it go?
I found SMB transfers at a sustained rate (no pausing or stopping) and its fast
Does SMB use the buffers in a different way than AFP?

any ideas on how i can test the speed of the disks?

they're all sata2 samsung 400gb in raid 5 array so should be reasonable?


is the megaraid controller the limiting factor? anyone else have throughput problems?

Strange!
I never had the problem that my transfers pauses. But today it started to happen (25MB/sec 8sec 0,0MB/sec 4sec).
Has anybody a solution?

I'm using it.