Howdy, Stranger!

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


When will BuyVM update pricing?
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.

When will BuyVM update pricing?

Time goes by and what in 2010 were competitive low prices, now are something between standard and higher than average prices. I think BuyVM isn't LowEnd anymore. Some time ago, when I needed to buy a VPS for a project I had waited BuyVM to restock, now I simply don't even take that into consideration. They were the firsts to offer a $15/year 128MB VPS, but their pricing is becoming more and more outdated. Why I would pay $12.95/month for a 1GB VPS if I can have the same specs with the same performance and reliability for less than $5?

What do you think?

«134

Comments

  • LeeLee Veteran

    Maybe they have just outgrown you?

  • When will BuyVM update pricing?

    Probably when they can't sell their servers anymore. If every stock gets sold out quickly, why would they lower their prices?

  • DamianDamian Member

    @albertdb said:
    Why I would pay $12.95/month for a 1GB VPS if I can have the same specs with the same performance and reliability for less than $5?

    More like, why would you bother worrying about BuyVM's prices, when this less-than-$5 provider apparently offers the same performance and reliability?

    We're no longer in the race to the bottom either, but we still can't keep things in stock at our prices.

  • jarjar Patron Provider, Top Host, Veteran

    @gsrdgrdghd said:
    Probably when they can't sell their servers anymore. If every stock gets sold out quickly, why would they lower their prices?

    Some would say this means prices need to go up.

  • albertdbalbertdb Member
    edited May 2013

    @Damian said:
    More like, why would you bother worrying about BuyVM's prices, when this less-than-$5 provider apparently offers the same performance and reliability?

    Because I have some VPSs with them since long time ago and until I recently tried other providers I though that noone offered their performance. Now I'm seriously considering cancelling all the BuyVM VPSs I have because I think I'm paying for more than what I'm getting.

    It's not about lowering the prices with no reason, but improving the service. What about a 50% increase in RAM? Things like that make customers happier.

  • It all comes down to personal preference but I think the service you will get at BuyVM exceeds other providers at this current time. All places have their downsides and everyone is entitled to their opinion.

  • @Liam said:
    Their performance is stellar

    It is not so stellar. A quick I/O test evidences this:

    Another provider:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    1073741824 bytes (1.1 GB) copied, 9.80341 s, 110 MB/s

    BuyVM1:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    1073741824 bytes (1.1 GB) copied, 10.7236 s, 100 MB/s

    BuyVM2:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    1073741824 bytes (1,1 GB) copiados, 10,8422 s, 99,0 MB/s

    10MB/s is a little difference, but in the price there is a big difference.

  • @Jack said:
    albertdb

    BuyVM has other features not just 'disk IO'

    It's just an example.

  • Getting fairly good disk IO here:

    BuyVM LV:
    [root@very ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync; unlink test
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 4.92574 s, 218 MB/s
    
    BuyVM NY:
    [root@reputation ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync; unlink test
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 4.06877 s, 264 MB/s
    

    More to the point, BuyVM has great network speeds and support, not to mention the fact that they're miles ahead of garden-variety providers with free backup space + free DNS hosting (though the latter is becoming more prevalent) and cheap add-ons like offloaded SQL and DDoS protection.

  • albertdbalbertdb Member
    edited May 2013

    @ihatetonyy said:
    More to the point, BuyVM has great network speeds

    BuyVM:~# wget -O /dev/null http://cachefly.cachefly.net/100mb.test
    --2013-05-26 00:52:55--  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 4.05M/s   in 17s
    
    2013-05-26 00:53:12 (5.95 MB/s) - `/dev/null' saved [104857600/104857600]
    Another provider:~# wget -O /dev/null http://cachefly.cachefly.net/100mb.test
    --2013-05-26 02:52:12--  http://cachefly.cachefly.net/100mb.test
    Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175
    Connecting to cachefly.cachefly.net (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 24.7M/s   in 4.5s
    
    2013-05-26 02:52:17 (22.4 MB/s) - `/dev/null' saved [104857600/104857600]
    

    As before, it's just an example. What I'm saying is that although their services are good, the product as a global doesn't justify it's price nowadays.

  • @Jack said:
    Are you using OVZ or KVM?

    OVZ in both providers.

  • RobertClarkeRobertClarke Member, Host Rep

    It's funny, because in an interview with @serverbear, @francisco talked exactly about how the disk IO is what most "cheapy" LETers consider to be one of the main and only performance factors with a VPS company.

  • @Jack said:
    You can't go and use all of

    What are you trying to say? Of course you cannot abuse the resources in both providers, but we are here talking about performance/price ratio, not about if it is enough speed for normal operation, which of course it is.

  • FranciscoFrancisco Top Host, Host Rep, Veteran

    I/O is something we're having to go back and improve alas. Our initial builds with the cache drives aren't working as well as they should and we've been able to find where the breaking points on the software end is. We're already discussing doing a Vegas wide upgrade with an added RAID10 of SSD's in each node to tally that up. We originally used 64GB SSD's since we expected some of the blocksize limits we put in place would keep the ssd's nice and happy. Alas, there's certain cases where we're actually maxing out he IO on them which is crazy.

    There's a few 128MB nodes that only have 4 disk RAID10's. These were put in place because people were demanding them so much and we were too tight on power to put large nodes in. Those gives anywhere from 150 - 200MB/sec when idle and probably 100 - 140MB/sec on average.

    There's a lot of upgrades and new features on the way that'll be in place within the next 2 months. Our new panel will be out by the middle of next month which will pave way for the novel worth of features I wrote in a thread a few months back.

    The brocade we had purchased to replace the FreeBSD box we got right now is actually being returned since it doesn't fit our needs anymore. A new router has already been ordered and we will be installing that the week of/after the panel goes in.

    A complete overhaul to our I/O is likely to come about in August when Anthony & I make a trip to Vegas to complete all of that.

    And finally, It's looking like we've finally found a 2.6.32 kernel that is stable enough for us. We have a few production nodes running it without issue so fingers crossed to that going in place.

    There has already been discussions about fiddling with our plans some, both with prices and resources. That isn't finalized though and won't be till the end of June most likely.

    We aren't being lazy or just sitting back, we simply got postponed on certain parts for far longer than I had hoped :(

    Francisco

  • @Francisco said:
    And finally, It's looking like we've finally found a 2.6.32 kernel that is stable enough for us.

    image

  • albertdbalbertdb Member
    edited May 2013

    @RobertClarke said:
    It's funny, because in an interview with serverbear, francisco talked exactly about how the disk IO is what most "cheapy" LETers consider to be one of the main and only performance factors with a VPS company.

    BuyVM

    --------------------------------------
    |        PHP BENCHMARK SCRIPT        |
    --------------------------------------
    Start : 2013-05-26 01:17:38
    Server : @
    PHP version : 5.3.3-7+squeeze15
    Platform : Linux
    --------------------------------------
    test_math                 : 5.327 sec.
    test_stringmanipulation   : 5.503 sec.
    test_loops                : 2.941 sec.
    test_ifelse               : 2.275 sec.
    --------------------------------------
    Total time:               : 16.046 sec.

    Another provider

    --------------------------------------
    |        PHP BENCHMARK SCRIPT        |
    --------------------------------------
    Start : 2013-05-26 01:18:31
    Server : @
    PHP version : 5.3.10-1ubuntu3.6
    Platform : Linux
    --------------------------------------
    test_math                 : 1.510 sec.
    test_stringmanipulation   : 1.529 sec.
    test_loops                : 1.134 sec.
    test_ifelse               : 0.910 sec.
    --------------------------------------
    Total time:               : 5.083 sec.

    Again, just an example.

  • @Francisco said:
    There has already been discussions about fiddling with our plans some, both with prices and resources. That isn't finalized though and won't be till the end of June most likely.

    Happy to hear that. I will wait until then to decide.

  • FranciscoFrancisco Top Host, Host Rep, Veteran

    You should compare CPU Mhz in that PHP test as well.

    It's likely that host is using E3 CPU's which are 3.2Ghz each compared to the 2Ghz - 2.5Ghz burst that ours do :) You'll notice we rarely suspend people from abuse and rather just put monitoring systems in place to help with things. If we were using E3's for our OVZ's we'd have to bust chops a heck of a lot more

    @albertdb said:
    Happy to hear that. I will wait until then to decide.

    Sure. I got some changes in mind but for now my #1 priority is the panel and our router. We were game to put the brocade in weeks ago in LV but that all went to hell when our needs changed on a dime.

    Francisco

  • @Jack said:
    One's running ubuntu one running debian and different PHP versions.

    Just in case it matters:

    BuyVM2

    --------------------------------------
    |        PHP BENCHMARK SCRIPT        |
    --------------------------------------
    Start : 2013-05-25 23:27:02
    Server : @
    PHP version : 5.3.10-1ubuntu3.6
    Platform : Linux
    --------------------------------------
    test_math                 : 4.554 sec.
    test_stringmanipulation   : 4.840 sec.
    test_loops                : 2.950 sec.
    test_ifelse               : 2.255 sec.
    --------------------------------------
    Total time:               : 14.599 sec.
  • @Francisco said:
    Our new panel will be out by the middle of next month which will pave way for the novel worth of features I wrote in a thread a few months back.

    A complete overhaul to our I/O is likely to come about in August when Anthony & I make a trip to Vegas to complete all of that.

    And finally, It's looking like we've finally found a 2.6.32 kernel that is stable enough for us. We have a few production nodes running it without issue so fingers crossed to that going in place.

    The most important thing is I trust the service that @Francisco has provided for the last 2+ years. :)

  • albertdbalbertdb Member
    edited May 2013

    @Francisco said:
    You should compare CPU Mhz in that PHP test as well.

    It's likely that host is using E3 CPU's which are 3.2Ghz each compared to the 2Ghz - 2.5Ghz burst that ours do :) You'll notice we rarely suspend people from abuse and rather just put monitoring systems in place to help with things. If we were using E3's for our OVZ's we'd have to bust chops a heck of a lot more

    Francisco

    It's an E3, but that CPU doesn't have 3x performance per core of a L5638 but 2x. Both VPSs have 1 vCore, so at maximum it should take 11 seconds in the L5638.

  • @Jack said:
    who's this other provider? I'm interested to know.

    I'm not here to do advertising, sorry.

  • I think there's a difference here. You can't just compare like that. There's also trust, reliability, stability and how things are handled when things break.

    There are a lot of hosts out there that are cheap because they don't care about healthy margins to stay afloat. A lot of them close down quickly. A lot of them also aren't run by a proper team with insurance, taxes and company structure.

    Price vs performance isn't the only thing to consider in a lowend world.

  • shovenoseshovenose Member, Host Rep

    A ShoveHost VPS will have faster IO than a BuyVM but who do you trust. Unfortunately for me, you're probably gonna pick the BuyVM. Oh, and I'm more expensive. If you don't like/want the service don't order it, don't complain in public.

  • risharderisharde Patron Provider, Veteran
    edited May 2013

    I do understand how @albertdb is feeling here since i do feel a little like this. I always want more from buyvm And primarily because i expect the best from @francisco

  • @concerto49 said:
    Price vs performance isn't the only thing to consider in a lowend world.

    I know that and it's not the only two things I'm taking into account.

    Since @Francisco said that they are going to update packages, I think it's pretty obvious that their prices are a bit high nowadays.

  • FranciscoFrancisco Top Host, Host Rep, Veteran

    @albertdb said:
    It's an E3, but that CPU doesn't have 3x performance per core of a L5638 but 2x. Both VPSs have 1 vCore, so at maximum it should take 11 seconds in the L5638.

    You have a lot more cache on the E3 which helps a lot. It's also quite possible that IO is affecting it some, it comes down to caches.

    You're welcome to ticket and I'll bench it against our internal E3 node and compare :)

    Let me make it clear, I'm always looking to improve our offerings and the many things we've done over the years should prove that.

    Francisco

  • FranciscoFrancisco Top Host, Host Rep, Veteran

    @albertdb said:
    Since Francisco said that they are going to update packages, I think it's pretty obvious that their prices are a bit high nowadays.

    Honestly? We're 2 years behind where we want to be with the company due to the falling out we had with our previous panel developers.

    Many of the things we've put in place already should have been done a long time ago so we've had to put stop gaps in place. Some of them have worked fine while others have shown they just aren't a great idea.

    With that being said, if you're after more features and now just bottom bin pricing, within the next 6 months our feature set will be as strong, if not stronger, than Linodes and far further ahead than digitaloceans, all without the higher price.

    Features like affordable hosted load balancers, anycast IP addresses, floating IP addresses between instances, and many other things, are all on the way and some are already partially coded.

    Don't get me wrong, on the hardware end of things we have points to improve and we are 100%, but also realize there's far more coming from us than just that :)

    Thanks for the interest,

    Francisco

  • @Francisco said:
    You have a lot more cache on the E3 which helps a lot.

    12MB Cache L5638 vs 8MB Cache E3, I've just checked it.

    You're welcome to ticket and I'll bench it against our internal E3 node and compare :)

    The script can be found here: http://www.php-benchmark-script.com/

    As per your other post, @Francisco, glad to hear that.

  • FranciscoFrancisco Top Host, Host Rep, Veteran

    the cache doesn't work like that on the L5638's alas :( Far as I know the cache gets split up in some wonky way so you end up with like 2MB per core or something.

    Anyways, i'll give your test a run. It's very much possible that there's new flags the E3's carry that help with it. I know back in the day Tim & I went through quite a bit of testing on L5420's and E3's and found some big differences.

    Francisco

Sign In or Register to comment.