[Littlemaster]

Good, then can you submit the output of these commands:

Code: (Select All)

lsblk  --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR,SUBSYSTEMS"
#
lshw -class storage -class disk

Code: (Select All)

# lsblk  --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR,SUBSYSTEMS"
lsblk: unknown column: SUBSYSTEMS

Code: (Select All)

# lsblk  --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR"
lsblk:  --output: not a block device
lsblk: NAME,KNAME,MODEL,HCTL,SIZE,VENDOR: not a block device

Code: (Select All)

# lshw -class storage -class disk
 *-ide
      description: IDE interface
      product: 82371SB PIIX3 IDE [Natoma/Triton II]
      vendor: Intel Corporation
      physical id: 1.1
      bus info: pci@0000:00:01.1
      logical name: scsi0
      version: 00
      width: 32 bits
      clock: 33MHz
      capabilities: ide isa_compat_mode bus_master emulated
      configuration: driver=ata_piix latency=0
      resources: irq:0 ioport:1f0(size=8) ioport:3f6 ioport:170(size=8) ioport:376 ioport:c640(size=16)
    *-disk
         description: ATA Disk
         product: QEMU HARDDISK
         physical id: 0.0.0
         bus info: scsi@0:0.0.0
         logical name: /dev/sda
         version: 2.5+
         serial: QM00001
         size: 100GiB (107GB)
         capabilities: partitioned partitioned:dos
         configuration: ansiversion=5 logicalsectorsize=512 sectorsize=512 signature=000b136f
 *-pnp00:03
      product: PnP device PNP0700
      physical id: 4
      capabilities: pnp

Hope we will reach to a conclusion soon.

[fChk]

Code: (Select All)

# lsblk  --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR,SUBSYSTEMS"
lsblk: unknown column: SUBSYSTEMS

Code: (Select All)

# lsblk  --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR"
lsblk:  --output: not a block device
lsblk: NAME,KNAME,MODEL,HCTL,SIZE,VENDOR: not a block device

That's weird!.. Normally, you should get something like this:

Code: (Select All)

[root @ fChk ~]# lsblk --output "NAME,KNAME,MODEL,HCTL,SIZE,VENDOR,SUBSYSTEMS"
NAME               KNAME MODEL            HCTL        SIZE VENDOR   SUBSYSTEMS
sda                sda   QEMU HARDDISK    0:0:0:0     100G ATA      block:scsi:pci
├─sda1             sda1                                 1G          block:scsi:pci
└─sda2             sda2                                99G          block:scsi:pci
 ├─cl_static-root dm-0                              57.9G          block
 └─cl_static-home dm-1                              41.1G          block
sr0                sr0   QEMU DVD-ROM     0:0:1:0    1024M QEMU     block:scsi:pci

Where the block device subsystem is clearly of a 'block :scsi:pci' type.

.. May be you should try:

Code: (Select All)

lsblk

Code: (Select All)

# lshw -class storage -class disk
 *-ide
      description: IDE interface
      product: 82371SB PIIX3 IDE [Natoma/Triton II]
      vendor: Intel Corporation
      physical id: 1.1
      bus info: pci@0000:00:01.1
      logical name: scsi0
      version: 00
      width: 32 bits
      clock: 33MHz
      capabilities: ide isa_compat_mode bus_master emulated
      configuration: driver=ata_piix latency=0
      resources: irq:0 ioport:1f0(size=8) ioport:3f6 ioport:170(size=8) ioport:376 ioport:c640(size=16)
    *-disk
         description: ATA Disk
         product: QEMU HARDDISK
         physical id: 0.0.0
         bus info: scsi@0:0.0.0
         logical name: /dev/sda
         version: 2.5+
         serial: QM00001
         size: 100GiB (107GB)
         capabilities: partitioned partitioned:dos
         configuration: ansiversion=5 logicalsectorsize=512 sectorsize=512 signature=000b136f
 *-pnp00:03
      product: PnP device PNP0700
      physical id: 4
      capabilities: pnp

Hope we will reach to a conclusion soon.

There you go!.. VPS-1 uses QEMU's emulated IDE interface as its storage controller (ata_piix.) Thus, contrary to what I was assuming, there is no VirtIO for your disk I/O virtualization!!.. It's just plain old IDE emulation!..

In other words, my first argument is no longer valid BUT the second one is still valid!.. As whatever storage is used at the host level there must be some form of RAID setup that can explain why your virtual disk buffered write speed is in ~300 MB/s (assuming the HDD storage type.) In which case, the rotational mode is derived from the RAID controller and not the disk drives per se.

Conclusion:
We're still in the dark :-) As I've already predicted in the first post. But it should be something like this, assuming the existence of a software RAID:
> HDD--SW-RAID--[-QEMU/IDE-Virtual-disk]
OR
> SSD--SW-RAID--[-QEMU/IDE-Virtual-disk]

Doing disk I/O benchmark tests may point to one case or the other but only if that SW-RAID is not too performant!

Having a highly performant RAID setup in the middle -especially HW-RAIDs- tend to blur the difference from a bench-marking perspective.