New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
For me it's less that I actually need that high of an I/O and more of I like to see that the providers take pride in their service to give you the best overall performance. Also, that they don't overload their nodes to the point where things seem so sluggish.
I don't really know if its good or bad
/tmp is a tmpfs so that doesn't really count
try from /root
Francisco
Oh i always thought /tmp was just located on the normal file system.
Anyway here you go with /root/, not much of a difference though
Are you not mistaking that with /dev/shm ?
+1
But, not always. For example my vserver at edis allocates 16MB for /tmp in RAM
That too, but i'm seeing it a lot more instances where /tmp is a tmpfs as well (debian, etc).
Francisco
ioping shows that isn't the case....
You're assuming that the OVZ devs actually coded something correctly
Either way, nice results!
Francisco
InceptionHosting
[root@mars ~]#dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 8.97836 seconds, 120 MB/s
[root@mars ~]#
XenSmart
root@uranus:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 10.5485 s, 102 MB/s
root@uranus:~#
Thehosthouse 256MB KVM
--- . ioping statistics ---
10 requests completed in 9037.7 ms, 416 iops 1.6 mb/s
min/avg/max/mdev = 0.5/2.4/4.4/1.2 ms
I'd say for the price, under 3 pounds a month, this is a keeper!
I'm thinking iops and stable latency to storage medium is more important than write speed, so how about we change this test to "ioping . -c 10", would be interesting to see what kind of iops people get with their vpses .
jollyworkshosting.com
Tiny Two package
Speeds differ.
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 39.6021 seconds, 27.1 MB/s
or
1073741824 bytes (1.1 GB) copied, 28.5103 seconds, 37.7 MB/s
or
1073741824 bytes (1.1 GB) copied, 45.1631 seconds, 23.8 MB/s
RamHost (UK-1, vz69)
vpn:~/ioping-0.6# ./ioping -c 5 /root/
4096 bytes from /root/ (ext3 /dev/sda1): request=1 time=0.4 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=2 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=3 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=4 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=5 time=0.4 ms
--- /root/ (ext3 /dev/sda1) ioping statistics ---
5 requests completed in 4003.0 ms, 2296 iops, 9.0 mb/s
min/avg/max/mdev = 0.4/0.4/0.5/0.1 ms
ChicagoVPS (1GB Package, Chicago VPS14)
[root@server3 ioping-0.6]# ./ioping -c 5 /root/
4096 bytes from /root/ (simfs /dev/simfs): request=1 time=22.0 ms
4096 bytes from /root/ (simfs /dev/simfs): request=2 time=18.3 ms
4096 bytes from /root/ (simfs /dev/simfs): request=3 time=16.4 ms
4096 bytes from /root/ (simfs /dev/simfs): request=4 time=0.2 ms
4096 bytes from /root/ (simfs /dev/simfs): request=5 time=19.0 ms
--- /root/ (simfs /dev/simfs) ioping statistics ---
5 requests completed in 4081.2 ms, 66 iops, 0.3 mb/s
min/avg/max/mdev = 0.2/15.2/22.0/7.7 ms
AsuraHosting (VPS-256, Master Node)
root@ea:~/ioping-0.6# ./ioping -c 5 /root/
4096 bytes from /root/ (simfs /dev/simfs): request=1 time=10.0 ms
4096 bytes from /root/ (simfs /dev/simfs): request=2 time=0.2 ms
4096 bytes from /root/ (simfs /dev/simfs): request=3 time=0.2 ms
4096 bytes from /root/ (simfs /dev/simfs): request=4 time=0.2 ms
4096 bytes from /root/ (simfs /dev/simfs): request=5 time=0.2 ms
--- /root/ (simfs /dev/simfs) ioping statistics ---
5 requests completed in 4016.5 ms, 459 iops, 1.8 mb/s
min/avg/max/mdev = 0.2/2.2/10.0/3.9 ms
vpn:~/ioping-0.6# ./ioping -c 5 /root/
4096 bytes from /root/ (ext3 /dev/sda1): request=1 time=0.4 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=2 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=3 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=4 time=0.5 ms
4096 bytes from /root/ (ext3 /dev/sda1): request=5 time=0.4 ms
--- /root/ (ext3 /dev/sda1) ioping statistics ---
5 requests completed in 4003.0 ms, 2296 iops, 9.0 mb/s
min/avg/max/mdev = 0.4/0.4/0.5/0.1 ms
And we have a winner!
I bet this one works so much better than any BuyVM hitting 500MB/s+ sequential write but only gives couple hundred iops.
I'm sorry, what?
best "what" of the week..
Burst.net $5.95/month 512mb OpenVZ VPS in Scranton
Hostigation $30/year KVM VPS in LA:
SecureDragon $16/year 128mb Xen
Home server (no raid, almost no load)
AlienVPS 128mb (my only LEB so far)
4096 bytes from /root/ (simfs /vz/private/5175): request=2 time=0.2 ms
4096 bytes from /root/ (simfs /vz/private/5175): request=3 time=0.2 ms
4096 bytes from /root/ (simfs /vz/private/5175): request=4 time=0.3 ms
4096 bytes from /root/ (simfs /vz/private/5175): request=5 time=0.2 ms
5 requests completed in 4002.3 ms, 4337 iops, 16.9 mb/s
min/avg/max/mdev = 0.2/0.2/0.3/0.0 ms
By the way, how do you quote properly?
Buyvm:
Misterhost.de
Server4You (not really lowend)
Cripperz.sg
I guess misterhost.de wins
Highlite some text in a post, and a little "Quote" link-in-a-box will appear near your cursor.
What about @gsrdgrdghd's? :O!
Yeah, that's pretty damn odd. It's worth giving it a second run... threads like these are murder. Someone mentions testing, and suddenly everyone slams the nodes <_<
Exactly. And at the end of the day it's all irrelevant. 'dd' and 'ioping' scores correlate more closely with phases of the moon than with real-world performance.
I bet he's on the older setups. The older setups are very hit/miss for that all. If he isn't then i'd for sure like to see a ticket.
Francisco
@Francisco:
According to Stallion: Node 128 (node38)
Hrm, nope, that's an upgraded one. I'm seeing around 1500 iops on the HN side so it's possible CFQ is just being cheap to you, but we're for sure strain testing .32.
There's a user ripping duplicity, wish this box was running .32 so I could see his IOPS
Francisco
Well i sometimes get better results too:
I'm using the server for OpenStatus monitoring only, so IOPS isn't important for me
But we really shouldn't turn this thread into a BuyVM support thread
Try using ./ioping -c 5 /root/somefile ; you'll get different results. I haven't look into how ioping works, but pointing to a directory or a file seems to make a difference
4096 bytes from /root (ext3 /dev/xvda1): request=1 time=0.2 ms
4096 bytes from /root (ext3 /dev/xvda1): request=2 time=0.2 ms
4096 bytes from /root (ext3 /dev/xvda1): request=3 time=0.2 ms
4096 bytes from /root (ext3 /dev/xvda1): request=4 time=0.2 ms
4096 bytes from /root (ext3 /dev/xvda1): request=5 time=0.2 ms
--- /root (ext3 /dev/xvda1) ioping statistics ---
5 requests completed in 4001.7 ms, 4803 iops, 18.8 mb/s
min/avg/max/mdev = 0.2/0.2/0.2/0.0 ms
4096 bytes from /root/test (ext3 /dev/xvda1): request=1 time=0.4 ms
4096 bytes from /root/test (ext3 /dev/xvda1): request=2 time=0.4 ms
4096 bytes from /root/test (ext3 /dev/xvda1): request=3 time=0.4 ms
4096 bytes from /root/test (ext3 /dev/xvda1): request=4 time=0.3 ms
4096 bytes from /root/test (ext3 /dev/xvda1): request=5 time=7.3 ms
--- /root/test (ext3 /dev/xvda1) ioping statistics ---
5 requests completed in 4009.5 ms, 570 iops, 2.2 mb/s
min/avg/max/mdev = 0.3/1.8/7.3/2.8 ms
That's a repeatable difference
@gsrdgrdghd any difference if you escalate to root first and then run it (so don't use sudo).