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
I have one in NY and one in Amsterdam.
Traceroute to Amsterdam:
2 7 ms 9 ms 7 ms 10.134.24.1
3 * * * Request timed out.
4 13 ms 15 ms 12 ms 24-139-16-165.eastlink.ca [24.139.16.165]
5 12 ms 12 ms 12 ms on-sdbr-dr001.on.eastlink.ca [24.139.7.66]
6 17 ms 12 ms 13 ms 24-139-15-26.eastlink.ca [24.139.15.26]
7 19 ms 17 ms 21 ms 24-139-7-138.eastlink.ca [24.139.7.138]
8 18 ms 17 ms 17 ms xe-4-2-0.tor10.ip4.tinet.net [77.67.68.181]
9 119 ms 157 ms 117 ms xe-1-1-0.ams12.ip4.tinet.net [89.149.185.93]
10 118 ms 124 ms 125 ms serverstack-gw.ip4.tinet.net [141.136.96.18]
looks like I got luck with my first server I just created another one and it practically took forever for the server to be created and the DD is also terrible compared to the first server.
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 4.87485 seconds, 220 MB/s
Or your ISP, for some reason routing to Amsterdam via the US.
Either case, just post your trace, before expecting any meaningful answer from anyone.
That isnt terrible?
I got this in AMS:
--2013-01-16 14:50:31-- http://cachefly.cachefly.net/100mb.testResolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected.HTTP request sent, awaiting response... 200 OKLength: 104857600 (100M) [application/octet-stream]Saving to: '/dev/null'100%[======================================>] 104,857,600 33.1M/s in 3.0s2013-01-16 14:50:34 (33.1 MB/s) - '/dev/null' saved [104857600/104857600]16384+0 records in16384+0 records out1073741824 bytes (1.1 GB) copied, 1.99582 s, 538 MB/scompared to my first server it sure is
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.2116 s, 486 MB/s
I can sorta second this, albeit with real-world stats.
I deployed two servers, one in NY another in AMS to operate as Tor relays.
After 12 hours, NY was averaging at about 30mbit/sec whilst AMS struggled at ~1mbit/sec.
I've since killed the AMS server, however I might give it another go tonight.
It would probably be a good idea to wait a couple days before you complain or are concerned about performance; they are probably so busy right now!
Actually most people post something like 280ms i/o throughput speed is not soo good... I mean wtf
Damn thing gone crazy.
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.2116 s, 486 MB/s
Right about the comparison, but knowing that a normal vps is around 100mb/s I'm really happy with it.
Yes it's definitely more than acceptable.. Just did a DD and seems to be about the same as the second server now which I deleted so can't really know if that has degraded further or not.
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 3.82138 s, 281 MB/s
--update
Pretty odd thing, I just created a new droplet and I received the same IP that was from the server that I deleted hours before??
well here is the DD
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 6.59529 seconds, 163 MB/s
Here is mine:
dd if=/dev/zero of=iotest bs=64k count=16k conv=fdatasync && rm -fr iotest
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 3.37272 s, 318 MB/s
@taronyu
Pretty good! I think i will delete the second droplet again..
You guys are bitching about ~200/300 IO speed? Really? Come on now, stop trying to be so pretentious, you'll never, EVER need to transfer that much shit around your server.
Get over it, you'll never notice any difference even if it was 600MB/s.
Get over it, you'll never notice any difference even if it was 600MB/s.
Where am I bitching? I like it.
Show us some ioping results from the VPSs with "slow" dd tests.
ioping . -c 10
ioping . -R
ioping . -R -s 256k
Here you go: (No I don't got a slow server, just because I like to do it
)
root@server:~# ioping . -c 10
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=1 time=0.5 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=2 time=0.7 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=3 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=4 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=5 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=6 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=7 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=8 time=0.5 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=9 time=0.6 ms
4096 bytes from . (ext4 /dev/disk/by-label/DOROOT): request=10 time=0.4 ms
--- . (ext4 /dev/disk/by-label/DOROOT) ioping statistics ---
10 requests completed in 9007.4 ms, 1760 iops, 6.9 mb/s
min/avg/max/mdev = 0.4/0.6/0.7/0.1 ms
root@server:~# ioping . -R
--- . (ext4 /dev/disk/by-label/DOROOT) ioping statistics ---
7000 requests completed in 3000.0 ms, 3041 iops, 11.9 mb/s
min/avg/max/mdev = 0.0/0.3/6.1/0.1 ms
root@server:~# ioping . -R -s 256k
--- . (ext4 /dev/disk/by-label/DOROOT) ioping statistics ---
6041 requests completed in 3000.0 ms, 2512 iops, 628.0 mb/s
min/avg/max/mdev = 0.3/0.4/5.0/0.1 ms
Well, when some get 200 MB/s and some get 500 MB/s it is quite fair to raise questions why some customers get a bettter product than others (even though it might be pure chance) and demand "the same as the other guy got". I posted above a method of trying to get that even w/o bothering with their support (using 5-10 test VPSes).
If everyone got 200 MB/s, then sure, no reason to complain.
Why do you even care? If 200MB/s is enough for you, where's the issue?
No offence, but are you working for DigitalOcean? You seem very affected if someone expresses even slight criticism or just his or her opinion. We are here for discussion. And this is what we do.
@rm_ Long before I started my web hosting business (back in December 2011), I was a customer at Zerigo (was also a Linode customer at the time). Zerigo offers a SLA as well (they advertise that they do). Needless to say that my 2GB VPS at the time ($80/mo.) went down in flames for no reason an "hanged". I could not reboot it, rebuild it, nothing. So I submitted a ticket and it took the about 72 hours to reply (3 days). In their reply they candidly reminded me that my service was unmanaged and that I shouldn't expect a fast reply for them. And while it took them half a day to fix the issue after the reply, they created a 512MB instance for me... so that I could have something to play with. Needless to say that I never got any SLA credit, and when I brought it up they didn't want to hear about it. I was out of there as fast as I could. Customer service like that was one of the reasons why I got into web hosting.
The bottom line is this: 1) You get what you pay for, so don't expect more or less. 2) A SLA makes for a great advertising tool. 3) Realistically speaking, expect a SLA if you have purchased Enterprise Level service. 4) Because of rule #3 rule #1 applies :-)
Aaaand I decided. Prometeus it is. I need the CPU power
If you want to host a Call of Duty GameServer on DigitalOcean: 18 slots is max. for the 512MB/1CPU plan. Above that it has lagspikes. And with 18 slots, you're living on the edge. Mainly due to CPU power: load average: 0.90, 0.85, 0.86
If everyone got 200 MB/s, then sure, no reason to complain.
There are different nodes, they aren't all going to be on the same node...
So if you know that other people was getting over 400 MB/s while your only getting 200 MB/s you would be fine with that? really you wouldn't try to get something better?
it was also the server with with the 200 MB/s literally took 15 min to build at one time I really thought it was hanging, you know this node is too full.
No, I wouldn't, because I can suffice with 200MB/s. I don't attempt to be a special snowflake by having a better IO speed than someone else.
it was also the server with with the 200 MB/s literally took 15 min to build at one time I really thought it was hanging, you know this node is too full.
In any shared service you are going to get fluctuations in performance. If I was getting 200MB/s I would be fine with that.
As for 15 minutes to build, they are pushing this heavy and invited existing customers to rebuild their server to take advantage of the offer. I would assume the disks are being trashed heavily at the moment.
Great you stick with the fully loaded server then, all the better for me!
If I was anyone, I would wait at least a month for this new pricing plan to settle. Personally, I'm going to wait at least 6 months for some features to come in, like an iOS app and IPv6.
I'm not moving off my @ramnode. I got a great discount from them and their performance is better.
the lowest I tested was at 160 MB/s so yes this node was pretty full it literally got stuck when the server was being built so that kinda did not give a good impression also I would think that the bandwidth would be considerably congested as well