Via this WHT offer. According to the About section on their website, CHVPS is “owned and directly operated by Private Layer INC. Private Layer is a Panama Corporation providing hosting services since 2010. Services offered under CHVPS.COM are provided from our Swiss Data Center location”. They have some low end VPS to offer, starting from $3.98/month for “Cirrus 1 VPS Package”. Ordering link here.
- 256MB guaranteed/512MB burstable memory
- 20GB storage (on RAID10)
- 1TB/month data transfer on 1Gbps
- Virtuozzo
On their Infrastructure section, it states that their data center is located in Equinix Switzerland Josefstrasse location, and they run their fully redundant BGP network under AS51852. Interestingly Parallel’s Virtuozzo was used, rather than the usual OpenVZ/SolusVM combo. More interestingly it utilises the UBC model of guaranteed/burstable memory, rather than SLM that has been available on Virtuozzo for a while.
PrivateLayer’s domain was registered in March 2010, and CHVPS was registered near end of June this year, i.e. just a month ago. That also happens to be when HostRail raised its price by 60% and subsequently went to the dead pool, as Constantinos owed $1,000 to — PrivateLayer! Hopefully a service directly from the upstream would be more stable.
Related Posts:
- 5 Reasons Why You Want a Low End Box - May 26, 2021
- Dead Pool January 2012 - February 2, 2012
- exit(0); - January 19, 2012
This location has always intrigued me but I just can’t pull the trigger on it because it is a 100% Cogent bandwidth only network as far as I can tell. http://bgp.he.net/AS51852#_graph4 shows that all they are directly connected to is Cogent. So much for fully redundant bgp ?
On the first tab you can also see ReTN.net as a peer, maybe due to costs/arrangements it is normally not used, and kicks in only whenever Cogent is down?
What is more problematic however, is that they have no IPv6 peers other than Cogent.
And because of the Cogent vs HE.net peering dispute, their network will not be reachable from HE.net (and vice versa).
my traffic don’t go near cogent http://pastie.org/2284769 its all telia
Their website is hosted in Panama so tracing there is irrelevant. Their Switzerland location (where the VPSes are actually hosted) is Cogent only. I had a hostfail windows vps hosted at this location and its just cogent in and cogent out. Which isn’t necessarily bad but when you hear things like Equinix DC, Redundant BGP, “Own private AS”, you expect more than 1 budget bandwidth provider.
Sir,
We actually have a connection to Hurricane, but it is administratively down at the moment for an upgrade to It will be back up on Monday, and will be 10Gbps. Also, you can see our other peers in the below BGP table. We are also adding a 10Gbps connection to Tiscali on Monday.
Thanks for responding. I’ll check back and possibly order when I see those connections up :)
The IO is workable for a normal website IF it stays the same and do not degrade further
by the way, can I have the test ips?
Francisco, you have a VPS with every possible provider? ;-)
Not yet :-) I have active servers and VPSs located at ~100 different data centers around the world and I guess I used 200+ data centers in the last 3 years.
wget http://byte-unixbench.googlecode.com/files/unixbench-5.1.2.tar.gz tar -zxvf unixbench-5.1.2.tar.gz cd unixbench-5.1.2 make ./Run dhry2reg whetstone-double syscall pipe context1 spawn execl shell1 shell8 shell16 BYTE UNIX Benchmarks (Version 5.1.2) System: ch3.1-ms.net: GNU/Linux OS: GNU/Linux -- 2.6.32-028stab085.3 -- #1 SMP Mon Mar 21 19:28:50 MSK 2011 Machine: x86_64 (unknown) Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968") CPU 0: Intel(R) Xeon(R) CPU E5503 @ 2.00GHz (4000.2 bogomips) Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Inte l virtualization 21:10:01 up 16:35, 1 user, load average: 0.00, 0.00, 0.00; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Thu Jul 28 2011 21:10:01 - 21:35:56 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 14547077.3 lps (10.0 s, 7 samples) Double-Precision Whetstone 1874.7 MWIPS (10.0 s, 7 samples) Execl Throughput 2146.9 lps (29.8 s, 2 samples) Pipe Throughput 640168.7 lps (10.0 s, 7 samples) Pipe-based Context Switching 190638.9 lps (10.0 s, 7 samples) Process Creation 4226.7 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 1952.4 lpm (60.0 s, 2 samples) Shell Scripts (16 concurrent) 192.3 lpm (60.2 s, 2 samples) Shell Scripts (8 concurrent) 363.5 lpm (60.1 s, 2 samples) System Call Overhead 562623.3 lps (10.0 s, 7 samples) System Benchmarks Partial Index BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 14547077.3 1246.5 Double-Precision Whetstone 55.0 1874.7 340.9 Execl Throughput 43.0 2146.9 499.3 Pipe Throughput 12440.0 640168.7 514.6 Pipe-based Context Switching 4000.0 190638.9 476.6 Process Creation 126.0 4226.7 335.5 Shell Scripts (1 concurrent) 42.4 1952.4 460.5 Shell Scripts (16 concurrent) --- 192.3 --- Shell Scripts (8 concurrent) 6.0 363.5 605.8 System Call Overhead 15000.0 562623.3 375.1 ======== System Benchmarks Index Score (Partial Only) 496.3There are some questions about our network.
We currently have Cogent Communications and Hurricane Electric as providers. Hurricane Electric is currently administratively offline as we are upgrading this connection.
On Monday 08/01/11, we will be bringing Hurricane Electric back online, but with a full 10Gbps uplink. Also on that date we will be bringing online a 10Gbps connection to TINET (Tiscali).
To get a full picture of how the network will look moving forward, please run some tests on Tuesday 08/02/11.
Thanks! Private Layer /CHVPS
So after Tuesday, do you think there will be a difference in response time? if it will get better or something?
Correct. Our plan for choosing providers follows the following logic:
Cogent: Great connectivity to North America.
Hurricane: Great connectivity to Asia
Tiscali: Great connectivity to Europe
We are are hoping with these diverse providers, as well our European peering, to provide our customers with a very robust network for their hosting products under AS 51852.
How about some level 3 :)
yes I second Level3
Also can I get test ip?
Our Edge is: 46.19.136.1
Test Download: http://chtest.privatelayer.com/100MB_test.bin
We checked with Level3 about 3 months ago and they did not service Equinix Switzerland. I just put in another inquiry to them to see if they do.
It doesn’t look like either the HE.net or Tiscali connections are active yet, is there a new date for when this will happen?
Do you have any .ch (swiss) IP block or all IPs belongs to panama?
I mean 46.19.136.1 seems like RIPE IP but does customers get also .ch/RIPE IPs or panama IPs?
46.19.136.1 is from our RIPE allocation, and under our network AS51852. All CHVPS customers get IP addresses from our RIPE allocation.
Hi mr Prado;
Can you add some more payment methods? Like AlertPay, Webmoney or LibertyReserve?
Thanks
Sir,
We used Ubersmith DE 2.0 for our billing software. This software is currently only compatible with Paypal, and 2checkout.
We however do take ALERTPAY and CASHU, but have to process those orders manually. Just talk to sales and they can help you out if you wish to pay via the methods.
Thanks for interest!
Hello,
is it possible to run a vpn server on these boxes?
Do you have any plans to offer ipv6 addresses on these VPSes? Perhaps some nice multihomed native ipv6 once you
get the new carriers hooked up :) ?
Yes – look for IPv6 in about 30 days time.
Any update on level 3?
Will this offer then still be valid? (once native ipv6 will be available too?)
Does someone else has problems with order? I’m waiting for my VPS for more than 10 hours!
Did you get your VPS already? Sorry for the delay, our weekend support desk was a little overwelmed.
Would it be possible to have the VPS shut off automatically if the bandwidth limit is reached? I doubt I would reach anywhere near 1TB but would like to guarantee no bandwidth overage fees. Some providers are setup like this.
Unfortunately, our software (Ubersmith DE) does not support this. You can however easily view your usage in real time in the control panel.
Hope they will stay stable and the I/O will not go down… That’s exactly what I was looking for :)
Is it possible to set up a VPN with that VPS? (seen that somebody did ask that already but no answer)
Thanks :)
We have no policy against VPN servers. So the answer is yes.
Ok, great! And technically it’s possible to use tun & tap?
Thanks!
@Prado: I signed up for this on Friday (29/07/2011) and have still not heard back.
@Karl, has your vps been set up yet?
An email sent to support@chvps on 30/07 has not been answered yet.
Also https://support.privatelayer.com/ does not seem to resolve right now.
What’s up?
I’ve received my VPS details!
The support site was showing an SSL certificate warning in firefox, which was why it appeared not to load. This is probably due the expired certificate noted by chip a few comments below.
After ignoring the warning, I was able to log in and submit a ticket.
A few hours later I received my VPS login details by email
Before I sign up I need to know if IRC is allowed (Internet Relay Chat). I have read over the AUP and ToS and no mention of IRC is made anywhere.
Am sure they do, and am sure they allow torrent search engines. as i am aware of 2 that are hosted there :)
It looks like they’ve botched Firefox’s “secure” connection icon (the padlock) on the paypal log-in webpage by loading a GIF of the “Private Layer” logo over an insecure link (HTTP instead of HTTPS). This makes firefox show a “Warning: Contains Unauthenticated Content”.
I’d feel more comfortable with their administration skills if they would make the purchase secure without requiring the user to evaluate whether the SSL link is safe. Please fix this asap.
James, can the paypal-login page be fixed so that the browser indicates the payment is secure?
Yes – our developer is working on that right now also. Problem is an expired SSL certificate.
Hmm.. as far as I can see CHVPS is still single homed disregarding the fact that they promised tiscali and HE working since monday. I hope that they didn’t have in mind some monday in the middle of 2012…
We are in the datacenter right now working on these connections. You will see both of them up in the next 12 – 24 hours.
I am looking forward to it.
Both connections are up:)
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
31.7.59.102 4 57048 20156 280677 36872214 0 0 02:50:41 1
46.19.141.158 4 34023 129562 6167953 36872214 0 0 12w5d 1
77.67.90.145 4 3257 74599 229 36872255 2 0 00:32:24 358860
91.206.52.32 4 21232 146129 129332 36872255 0 0 00:33:34 15
91.206.52.45 4 44706 129300 129307 36872255 0 0 00:33:32 2
91.206.52.50 4 12399 142373 129332 36872255 0 0 00:33:29 9
91.206.52.56 4 20940 128438 129334 36872255 0 0 00:33:27 6
91.206.52.59 4 35406 141847 129314 36872255 0 0 00:33:54 2
91.206.52.63 4 31662 128672 129349 36872255 0 0 00:33:27 5
91.206.52.70 4 34288 142401 129335 36872255 0 0 00:33:27 10
91.206.52.71 4 34288 142246 129297 36872255 0 0 00:33:16 10
91.206.52.100 4 21232 146148 129332 36872255 0 0 00:33:34 15
91.206.52.155 4 12387 142295 129276 36872255 0 0 00:33:32 9
91.206.52.166 4 44919 129345 129327 36872255 0 0 00:33:52 5
91.206.52.179 4 49983 811804 774703 36872255 0 0 00:33:28 3
91.206.52.181 4 39912 132640 129331 36872255 0 0 00:33:30 149
91.206.52.185 4 21385 269006 258509 36872255 0 0 00:33:22 107
91.206.52.251 4 42476 1266011 258652 36872255 0 0 00:33:20 11402
149.6.176.169 4 174 9963047 129582 36872255 0 0 01:50:17 358997
216.66.84.201 4 6939 427361 3035 36872255 4 0 00:06:43 360098
Very low ping too!
1 <1 ms * <1 ms 46.105.103.253
2 1 ms <1 ms <1 ms rbx-g1-a9.fr.eu [178.33.100.25]
3 5 ms * 6 ms ams-1-6k.nl.eu [94.23.122.69]
4 6 ms 13 ms 6 ms 20gigabitethernet1-3.core1.ams1.he.net [195.69.1
45.150]
5 11 ms 17 ms 10 ms 10gigabitethernet1-1.core1.fra1.he.net [72.52.92
.93]
6 17 ms 17 ms 17 ms 10gigabitethernet1-1.core1.zrh1.he.net [72.52.92
.230]
7 17 ms 16 ms 16 ms private-layer-inc.gigabitethernet2-6.core1.zrh1.
he.net [216.66.84.202]
8 16 ms 16 ms 16 ms 31.7.58.98
9 16 ms 16 ms 16 ms http://www.chvps.net [31.7.58.117]
So is anyone experiencing 100% stable service from these guys?
I saw some rumblings on WHT that would suggest some minor issues, maybe they have been worked out though.
From our monitoring station located in Portugal:
Uptime 100,0000% [6 d 18 h 2 m]
Downtime 0,0000% [0 s]
Uptime&Downtime Total 6 d 18 h 2 m [=100% coverage]
Uptime Coverage Since 27-07-2011 20:51:35 [6 d 18 h 3 m ago]
Report Time Span: 27-07-2011 00:00:00 – 04-08-2011 00:00:00
Sensor Type: Ping (60 s Interval)
Probe, Group, Device: EMEA NOC-5 > Europe > Zurich 3
Uptime Stats: Up: 100 % [6d18h10m22s] Down: 0 % [0s]
Request Stats: Good: 99,99 % [9728] Failed: 0,01 % [1]
Average (Ping Time): 44 msec (*)
(*) Starting yesterday the latency increased from 44ms to 55ms.
Just to be fair, we have another CH VPS with EDIS and the latency is ~56ms.
Report Time Span: 15-06-2011 00:00:00 – 04-08-2011 00:00:00
Sensor Type: Port (60 s Interval)
Probe, Group, Device: EMEA NOC-5 > Europe > Zurich
Uptime Stats: Up: 99,803 % [49d12h39m8s] Down: 0,197 % [2h20m33s]
Request Stats: Good: 99,764 % [71298] Failed: 0,236 % [169]
Average (Available): 56 msec
Sounds good.
Still no problem, uptime and I/O wise?
Sensor Type: Ping (60 s Interval)
Probe, Group, Device: EMEA NOC-5 > Europe > Zurich 3
Uptime Stats: Up: 100 % [11d1h51m52s] Down: 0 % [0s]
Request Stats: Good: 99,994 % [15949] Failed: 0,006 % [1]
Average (Ping Time): 49 msec
hmm, pretty bad IO, it should be at least 25 MB/sec
Did you raised a ticket?
Disk I/O on the $3.98 plan is limited to 10.8MB/sec so you will never get over that.
Ah, explains. Thanks
“Joe Merit:
Disk I/O on the $3.98 plan is limited to 10.8MB/sec so you will never get over that.”
Actually the initial test (ran 11 days ago) the throughput was over 28MB/s
If you go into Virtuozzo Panel and look at the bootup log for your vps it says “Set up iolimit: 10485760” I assume this translates to 10.8MB/sec because that is exactly what I get when I do the test. EVERY TIME.
@Joe Merit
Very bad. This is not acceptable. I hope CHVPS fix this ASAP.
Thank you very much for the information. I really appreciate it.
I signed them, but not like its web site said 1 hour ensure the installation.
Yeah LowEndBox? really? looks a lot more like LowEndScam. U pay to get a crappy billing system where they suck you credit for nothing, no support, no reply at all, no VM, no instructions nowhere, no login info, no account…
I bet this fraud is gonna be down very soon!
Had a bad experience with them?
Can you tell us more?
Thx!
I got the Cirrus 1 plan yesterday, but since then I’ve received no email at all.
i managed to get my account details from the password recovery page, that led me to a billing system w/out any signs of account detail to access my VPS, opened a ticket then yesterday, no answer so far and chat is always offline. they states 12 hours to setup, it’s more likely 24 right now, but at least they could send an email
so just to be clear, have you received any welcome email from them?
How Im supposed to know the account details?
maybe Im missing something.
I finally got an email from them some 2 days later, stating the had an error with the Paypal script, reapplied payment and would get VPS up in 2 hours.
Apologies, our billing department is not open on the weekends. So were were able to catch the error today.
ok, just received account details by email, thanks!
How is it going with your VPS? Is the IO limit annoying for your usage?
:)
“Asim Zeeshan:
hmm, pretty bad IO, it should be at least 25 MB/sec
Did you raised a ticket?”
I just redo the test to answer Bob question and I’m surprised too but I’m not intending to open a ticket. This VPS was a test-drive for a future project.
Did they fix this incredible low “iolimit 10485760” finally or their vpses are still crippled?
No fix yet.
The newest Virtuozzo supports placing IO caps on a per VM basis so it’s quite possible they’re enforcing it to be no faster than that.
Francisco
“storage (on RAID10)” to me appears as though they want to say I/O will be fast.
With the 10MB/sec limit which is not advertised anywhere in BIG LETTERS prior to purchase, I conder this company and plan to be borderline fraud.
Yeah, that’s pretty low.
A limit might help preventing some heavy uses of the HDD though, but well, it’s way too low… :/
Must be written in BIG LETTERS priori to purchase IMHO.
for me is a test VPS, but with this HDD I/O cap it’s pretty useless, besides they didn’t write this “feat” anywhere. what do u think? it’s enough for a refund?
It was a bad move from their part. As I said before the first test I did the disk I/O was over 28MB/s. They changed the rules after the game started.
Yeah, that’s not really honest / respectful.
We will see in the next few days if somebody from there staff comes here to comment on this issue / if they decide to do something or not.
I wanted to order, but am waiting for an eventual change before ordering, or I’ll have to go for another provider..
Thanks for all the feedback (positive and negative)
I apologize for the confusion. The disk I/O limit is in place in order to allow each user a fair share of the servers disksystem resources. We obviously cannot allow one user to hog all the resources for all the other users. We were experimenting with a few limits, and I will admit that the 11 MB/s is a little low.
Effective immediately, the disk I/O policy for all our VPS servers are:
Cirrus Class Packages : 25MB/s
Cumulus Class Packages: 25MB/s
Strato Class Packages : 100MB/s
Any customers that are not happy with the new I/O policy can email billing@privatelayer.com. We will promptly refund your account with out any questions asked.
Hey, great news.
That’s still low but a little bit more usable :)
How many users do you fit in one machine?
Depends on the orders that we get and how much resources we promise for each server. Our main limiting factor is the RAM on the box, and normally this determines the amount of customers we can fit on the box.
It is well known the main limiting factor is I/O.
So your the expert on our VPS nodes now? How do you even know what hardware we are using and what the main limiting factor would be?
You are free to disagree with my opinions and my reasoning but no fallacious arguments please.
I don’t think this is great news. You have “up to” 25MB/s not “guaranteed” 25MB/s then just one VPS doing I/O at 100MB/s will impact everyone performance. Poor police, poor decision when many VPS providers offer disk I/O at 50-100-200MB/s at the same price point (e.g. EDIS).
Your argument doesn’t make any sense.
Are you for a low disk IO policy, or a high disk IO policy? You claim “Other providers allow disk I/0 at 40 – 100 – 200 MB/s”, but then this affects other users correct? Do you expect that we can give out a dedicated disk drive to each $4.00 VPS? Or maybe we should just lie to you like other providers and say “You have a dedicated 100MB/s IO, when you know its not true.
I am sorry that you are unhappy with our policies, they are meant to protect all the users of the box. If you require a 200MB/s dedicated disk IO, then I suggest that you go with a dedicated server, or go with another VPS provider that is going to lie to you about the resources you will have. We will gladly provide you a refund.
The fact is that with a lot of budget providers (edis is a good example, but it works with buyvm too even if their I/O is not as good as edis) I/O is not voluntarily limited, and you can IO faster than 25 MB/s….
Hey, is there anyone here who did order who can tell us how it has been for him/her?
Is it stable, how good is the connectivity..
Thanks!
@James Prado
Is video files allowed to be stored on these VPS’es ?
Finally took one VPS with them to test the thing for a future project.
IP in the range 31.7.63.128 – 31.7.63.159:
The VPS arrived came running CentOS 5, which is not what I ordered, and it seems like there is no way to automatically reinstall the system…
Surprisingly good I/O though – or i’m on a fresh node / another one than you guys and they didn’t setup the limit yet, or they did listen the community and remove the limit…
Will see in the future how it goes – if you have questions / want special testing, just ask!
Hum, I guess I was on a pretty new node…
I tried it for a month and was satisfied with everything except for their ridiculous
Disk I/O policy. I would not have signed up if I knew I was going to be limited to 10.8MB/sec of IO.
They improved this to 25MB/sec which is somewhat acceptable for a $4 plan but still not really
all that acceptable to me when a my OpenVZ & KVM vpses that are a lot cheaper than CHVPS will always
give me over 100MB/sec when I am accessing the disk.
However.. 25MB/sec on the lower plan and whatever they have now on their higher plan will probably
be fine for many people.
I meant to say my BuyVM OpenVZ and KVM plans.
Pfff, and now everyone believes that the minimal level of quality of service is standarized by BuyVM ¬_¬ crap
That’s not so bad as they are pretty decent for this money :P
Not by BuyVM Yomero, but by the Glorious Pony…
I would like to see more providers at their level.
It’s pretty sad that buyvm’s unmanaged support is better than many providers managed service around here.
I’m confused, is this a stab at us or?
While we might give limited helping hands in tickets, our community based support is top notch? I mean, we even have competing companies helping in channel just because they seem to enjoy it.
Francisco
Hey, I’m just idling waiting to troll bzImage
Its a little off topic as well, considering this is a CHVPS thread :)
Is to me Fran?
Nah, is a Big AXE!! j/k
Well, I mean, everybody is expecting to get the same performance with every provider and taking yours as a base. That is unfair (buyvmish old thread). Every provider has is own circunstances:
Datacenter
Location
Hardware
Staff
And while this particular provider is not good with their I/O cap, there are others around with pretty decent specs and everybody is trying to compare.
Enough,my english still sucks xD
Their broadband is often abuse, not stable. Often appear to slow.
Take a look in Power Panel, Then resources, If network Shaper is enabled ask them to disable it. I have found it limits you to around 4-5Mbps
If it has 5 M, will not say I four VPS all at the same time, at the same time slow unable to visit. I can’t each VPS, are beyond 5 M broadband use.
Power panel no longer accessible here. Can anyone else access theres?
It works for me.
Welcome to Parallels Power Panel.
Status: The Virtual Environment is running at the moment.
So far so good, quick provisioning, they “forgot” to add the second IP
I ordered, installed dnsonly and it was quick..
A vps cheaper than a shared hosting is a good deal, if you take into
account the setup, billing system etc it’s a steal!
They’re now migrating from Virtuozzo to OpenVZ/SolusVM on a new container, with a new ip.
With double storage and memory, now using vswap.
They are migrating all existing customers?
Yes, they are no longer using Virtuozzo
cpu down from 2133 Mhz to 800Mhz
Sirs,
We have been provisioning servers on Solus/OpenVM for about 3 months now. The only customers being moved are clients that are on older Virtuozzo Nodes. All the migration notices have been sent out now. If you did not get a notice, then you are already on a Solus/OpenVM node. If you have any doubts, please email us at support@privatelayer.com and we will clear it up for you immediately.
Sincerely,
James Prado
Private Layer INC
Weird, when did you send the notifications?
Bob,
They were sent over the last 48 hours. Just open a ticket with support@privatelayer.com and they will check for you.
James
Did find it, thanks!
I/O is quite bad though :/
well, instead of sending notification 5 days before shutting down, you had better notify a little earlier!!!!! ….Paid service shut down with 5 days of notification and all the migration work on the costumer! i dont have time to waste on this in the next 5 days…
Ouch…
James: do you have 200 customers per node on a SATA RAID 1?
I think I’ll ask for a refund if things don’t get better.
They limit io/speed this is well known and has been discussed. it aint a bad thing realy..
Wasn’t the limitation supposed to be around 10 or 20MB/s?
(At least that’s what I remember from the discution, my tests on the ‘old’ node were usually better than that :))
Just did a test on the new node i’m on
Not very good but still can be used – I don’t expect 120MB/s from them, but less than 1MB/s is pretty bad for those who intend to run a database server for example… (10 or 20MB/s sounds like a rather fair limit to me, and as everybody is not writing at the same time on the HDD & VPS nodes usually have RAID 10, it can be the ‘price’ for a stable I/O for everybody)
@ John: maybe they added a lot of customers in your (our?) nodes who are testing the box / untarring their backups?
Will see how it goes in the future :)
This node, that we moved a majority of the customers, has currently has a raid 10 setup with 4 drives (western digital black series). We moved about 55 customers to this node and will leave it at that number.
I will check with support about the drive limitations, and post back into this thread for you.
We have a lot of customer moving into this box at the time, and I think the poor tests are due to the fact everyone is moving content into the box, testing it, all at the same time.
Give it some time to settle in, and I am sure that the speeds will be better than what you had on the old box. If not, we will gladly move you to another node upon request.
Ok, great.
Hope it will be better soon…
Sounds kinda low to me :p
I signed up for this and I think maybe something is wrong? their 100mb.test file on their site is getting 70 KB a second, and a rsync server to server connection is getting the same. It’s not my connection, I am in USA with a 25 Mb/s broadband.
If this is their normal network avoid at all costs, website is so slow.
Forgot to copy everything. Here it is:
[root@04001011608QA ~]# wget freevps.us/downloads/bench.sh -O - -o /dev/null|bash CPU model : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz Number of cores : 1 CPU frequency : 638.858 MHz Total amount of ram : 256 MB Total amount of swap : 256 MB System uptime : 0 min, Download speed from CacheFly: 3.42MB/s Download speed from Linode, Atlanta GA: 2.68MB/s Download speed from Linode, Dallas, TX: 93.3KB/s Download speed from Linode, Tokyo, JP: 47.0KB/s Download speed from Linode, London, UK: 10.4MB/s Download speed from Leaseweb, Haarlem, NL: 47.0MB/s Download speed from Softlayer, Singapore: 37.7KB/s Download speed from Softlayer, Seattle, WA: 217KB/s Download speed from Softlayer, San Jose, CA: 68.9KB/s Download speed from Softlayer, Washington, DC: 91.3KB/s I/O speed : 598 kB/s [root@04001011608QA ~]# free -m total used free shared buffers cached Mem: 256 37 218 0 0 13 -/+ buffers/cache: 24 231 Swap: 256 13 242 [root@04001011608QA ~]# 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, 27.5115 seconds, 39.0 MB/s [root@04001011608QA ~]# wget http://cachefly.cachefly.net/100mb.test -O /dev/null --2012-05-06 20:51:54-- http://cachefly.cachefly.net/100mb.test Resolving cachefly.cachefly.net... 205.234.175.175 Connecting to cachefly.cachefly.net|205.234.175.175|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 6.86M/s in 14s 2012-05-06 20:52:08 (7.13 MB/s) - `/dev/null' saved [104857600/104857600]