Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Rackspace Cloud Server Benchmark
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.

Rackspace Cloud Server Benchmark

shovenoseshovenose Member, Host Rep
edited February 2013 in General

First of all, dd...

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, 9.04251 s, 119 MB/s

Will post the serverbear in a moment :)

Comments

  • Will be interesting to see what you get. Never bothered to benchmark there.

  • shovenoseshovenose Member, Host Rep

    They bill hourly so it's not too expensive of a test.

  • Yup. Just never bothered.

  • shovenoseshovenose Member, Host Rep

    Benchmark completed disk test and bandwidth test. doing unixbench right now looks like.

  • We use several of these servers at work.
    Rackspace is average about disk performance.
    About the processors, really low compared to most providers.
    But the network is stable, and almost never goes down.
    Also we have 100% hardware uptime since ALWAYS.

    And well, is very very expensive.
    But they have a LOT of features in addition to VPS servers.

  • shovenoseshovenose Member, Host Rep
    edited February 2013

    @yomero that's how I feel about everything Rackspace... mediocre boring performance, but rock solid stability.

  • Wow.

    Well, they have an enforced CPU scheduling policy =/
    Unixbench is SO intense... and in my opinion, nobody must be using it.

    Also, unfortunately their new openstack based servers perform worse than their previous line. AFAIK you still can create one of these from the old panel.

  • shovenoseshovenose Member, Host Rep

    How can you tell that they have "an enforced CPU scheduling policy" ?

  • @shovenose said: How can you tell that they have "an enforced CPU scheduling policy" ?

    In my experience, you can use all the CPU available, but you get throttled when you use a lot and the server becomes slow... =/

  • @yomero said: We use several of these servers at work.

    Rackspace is average about disk performance.
    About the processors, really low compared to most providers.
    But the network is stable, and almost never goes down.
    Also we have 100% hardware uptime since ALWAYS.

    And well, is very very expensive.

    But they have a LOT of features in addition to VPS servers.

    Their support is also very responsive at least in my experience.

  • @jbxl said: Their support is also very responsive at least in my experience.

    Well, about this, I have a mixed opinion. Yes, the live chat is really a live chat, but some of these people doesn't have a clear idea of what they are doing, lol =(

  • Ah ok, Never used live chat... well maybe one time. I've used the e-mail support and had good response time. Knowledge is certainly not top tier but it was appropriate for the question asked.

  • @shovenose said: but rock solid stability.

    You got that right. I had 256MB box there and the longest uptime was almost 450d.

  • we use a set of dedicated Rackspace servers at my work. Expensive, yes. 100% uptime, yes. Support that actually can fix things over the phone, yes.
    In the end when you are running mission critical software it is worth it for the stability and support.
    Though I love Rackspace, for non critical services I would look else where.

  • KuJoeKuJoe Member, Host Rep

    Wow, Rackspace must be hurting if they are getting into the LEB market. :(

  • @KuJoe said: if they are getting into the LEB market

    Where it says that? o_O!

  • Sarcasm detector, you need one.

  • @Wintereise said: Sarcasm detector, you need one.

    I am Sheldon...
    Ok ok, I will excuse :P English isn't my main language :D

  • MonsteRMonsteR Member
    edited February 2013

    @shovenose said: First of all, dd...

    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, 9.04251 s, 119 MB/s

    Will post the serverbear in a moment :)

    Got a Softlayer cloud for $50 a month and it does about 187Mb/s really nice at softlayer

  • jhjh Member
    edited February 2013

    @MonsteR said: Got a Softlayer cloud for $50 a month and it does about 187Mb/s really nice at softlayer

    Did you get local or SAN storage? I got one with SAN and had 6 hours of unexplained downtime while a tech I wasn't allowed to speak to manually failed over my instance. I asked about credits and they pointed me to a large SLA I couldn't be bothered to read. $50 wasted.

  • shovenoseshovenose Member, Host Rep

    Softlayer is a damn ripoff.

  • @jhadley said: Did you get local or SAN storage? I got one with SAN and had 6 hours of unexplained downtime while a tech I wasn't allowed to speak to manually failed over my instance. I asked about credits and they pointed me to a large SLA I couldn't be bothered to read. $50 wasted.

    Well I am on there trial at this time, But I think the storage was local but seems really nice, There panel is nice as well some really cool features like you can SWIP an ip in seconds to your details. I noticed there support does lack a little dedication as I think they have about 40 windows loaded at the same time.

  • shovenoseshovenose Member, Host Rep

    @MonsteR but for the price Softlayer charges support shouldn't have 40 windows loaded at the same time!

  • Yeah they shouldn't but I can think of loads of places that can do stuff better and have chosen the cheaper route shame really but they do have good service.

  • KuJoeKuJoe Member, Host Rep

    @yomero said: Where it says that? o_O!

    I figured since somebody was posting benchmarks of a VPS on a forum specifically for LEBs this thread would be related to LEBs.

  • shovenoseshovenose Member, Host Rep

    @KuJoe just thought it was an interesting comparison to a LEB.

Sign In or Register to comment.