Page 1 of 4 1234 LastLast
Results 1 to 25 of 79
  1. #1

    Under $50 VPS plans compared

    From the guy who brought you, "Hosting? Read my tips" (http://www.webhostingtalk.com/showth...hreadid=309910), I now offer the community a little Excel spreadsheet I made to help me compare under $50 VPS plans.

    Actually, I created an in-depth breakdown of many attributes of plans from companys that have been mentioned here about 2 months ago, and JUST as I was about to upload it here, I accidently erased the file!!

    Now two months later, with all hope of recovery of my hardwork gone, I decided to spend the day and visit some sites, collect data, and post my comparison matrix.

    In a nutshell, I've ran my sites on virtual hosting, tried a dedicated server, and now feel that VPS might be right for me. I'm paying about $30 for my reseller acount now on http://acunett.com/, so hopefully I can find a plan under $40 that meets my needs. (Yeah, I don't think my sites warrant a dedicated server.)

    My comparison matrix might be handy to anyone else looking for a low-cost VPS plan. Though keep in mind companies are always (a) changing specs/prices (b) going out of business (c) having their support team taking rollercoaster rides.

    Perhaps with the help of you all, I can keep my file updated. So, if you see data which I need to add/edit to a company I listed, or a company that SHOULD be added, please let me know.

    Disclaimer: I know there are many many more VPS companys in the world, but I don't plan to list them all! I'll only added ones that this community reports on. Also, the companys listed may not be "the best", and some just "average." Again, I just did a search to find out what plans people were talking about, and then visited their sites.

    Anyhow, I hope this is useful to someone.

    Cheers,

    Carlos Camacho
    Attached Files Attached Files
    Last edited by Carlos; 02-08-2005 at 04:14 AM.

  2. #2
    Thanks for the effort, iam checking the file now
    You might want to add burst.net new vps offerings.

  3. #3
    Join Date
    Jan 2003
    Location
    Chicago, IL
    Posts
    6,957
    Noticed at least one correction, Equinix isn't a network.
    Karl Zimmerman - Founder & CEO of Steadfast
    VMware Virtual Data Center Platform

    karl @ steadfast.net - Sales/Support: 312-602-2689
    Cloud Hosting, Managed Dedicated Servers, Chicago Colocation, and New Jersey Colocation

  4. #4
    >burst.net
    Darn, they were on my list but I missed them. I'll need to add them in.

    >Equinix isn't a network.
    Indeed. I'll need to remove that.

  5. #5
    Join Date
    May 2004
    Posts
    398
    Thanks for the effort, I took a look at the file, looks pretty nice.

  6. #6
    Join Date
    Aug 2003
    Location
    Vancouver, BC
    Posts
    1,894
    I guess we didn't make the cut
    Gary Jones

    BlueFur.com - Canada Web Hosting

  7. #7
    Hello all,

    Thanks to a PM, I will add myriadnetwork.com, and re-upload the new comparision.

    >I guess we didn't make the cut
    Not really "a cut." Rather, my search of about four dozen "Which VPS?"-type threads didn't make me aware of your company. Like myriadnetwork.com, why not send a PM?

  8. #8
    Seems that the forum only gives is 15 minutes to edit a post. Hmpf.

    So here is the latest comparison.

    Added...

    monsterhosting.ca
    myriadnetwork.com
    Attached Files Attached Files

  9. #9
    It seems I can't connect to www.burst.net website. Not sure if they are down or it has to do with my network connection. If anyone has time, please visit their site and grab me the info I need for their best under $50 VPS plan.

    Cheers,

  10. #10
    Never mind, got through to burst.net.

    I'm in the process of updating my comparison, and also adding some new companies that I found via Google... names that perhaps don't pop up here so much if at all.

    Cheers,

  11. #11
    Darn, wish there was a way to edit posts, so I don't like a fool talking to myself!

    I've been thinking of the following comment which was posted my a member here...
    In the same sense, all VPS's should be slower than being on your run-of-the-mill shared server, assuming that server isn't overloaded. There's simply more software in between your applications and the hardware. VPS's should give you better minimum performance, but worse peak performance. Whether you'd ever notice the decrease is questionable, but if you moved from an overloaded shared server to a VPS, you should definitely see an increase in minimum/average performance.
    I'm on a shared account now (Reseller/Plesk). This comment has me a bit worried about moving to a VPS. Should it?

  12. #12
    Join Date
    Jun 2000
    Location
    Washington, USA
    Posts
    5,990
    Good luck on making the comparison chart. Looks like a good idea.

  13. #13
    Whew, my fingers are killing me from contacting companies!

    Below is the latest update to my comparison. I've updated some information from the first 1/2 dozen companies that have contacted me (thanks!), and added some non-WHT VPS companies (ie, they don't live here )

    To help answer my own question concerning the quote in my previous post, I've contacted at random a dozen people who are running vBulletin on VPS plans for their feedback.

    Finally, I should state that for other people looking at VPS plans, please remember that although "specs" are important, service, staff, company, ("little details") are just as important! (i.e. My comparision might focus on specs, but my final choice, as yours should, will take other factors into consideration.)

    Cheers,

    Carlos
    Attached Files Attached Files

  14. #14
    I'll be uploading a new version of my comparison in the next post. Before that, I'd like to share some comments I received from the the various companies I contacted. I have of course removed names. I hope by posting some of the information I was provided, it might help people who are looking at VPS plans to consider issues that they haven't thought of before...

    Here are the comments (which I found interesting) from various sales staff...

    We do not 'manage' your server (can anyone define what that even means?). However we work hard to provide great support to whatever requests we get. We operate a decent, honest business. We work hard to treat our customers well and provide a professional service -- we have been providing VPS services for a number of years and our server and support setup is very mature, stable and well-tuned.
    The bottom line is this - we want you as a customer "long" term. We will accomplish that by providing a good service, at a fair price, backed by friendly and knowledgeable support engineers. We have been in business over four years now and most of our new customers come by way of referrals.
    I won't bore you with the good and bad of capping each VE or not capping, you can read WHT for the pros and cons, but you'll see we consistently score in the top 2 or 3 on UnixBench testing. Providers that score poorly will pooh-pooh it, which is fine because customers eventually test our "Equal Share" configuration over a hard-cap and say "Wow!
    My suggestions for conducting a true, non-biased VPS comparison would be to purchase an account from each, and run a standard http request/cpu load/disk io benchmark on each and compare the results. For example, we can claim 200GB of bandwidth and dual 3GHz processors, but it is dependent on the oversale ratio each business strategy maintains. Raw numbers, although valuable and acurrate, paint a very small picture of the service you are likely to receive.
    If you plan to run a dynamic site (i.e. database driven), then 128MB for a VPS simply is not enough. You really need to go with at least 256MB, or your vBulletin and CMS will start to report problems.
    We don't feel a public forum is required; our support ticket system handles all our customer's needs.
    Virtuozzo based VPS will give you hands down the best performance, based on our extensive research and testing.
    When you outgrow a shared VPS server, you can upgrade to what we call a dedicated VPS server. Not many (any?) providers are offering this, and you face the same old problems of out-growing your current host, and having to search for a dedicated.
    We choose to be very honest and open with our clients, evidenced by our "no-hassle" guarantees. You can try our VPS for 30 days -- if you don't like it, we'll give you a 100% refund, including setup fees, no questions asked.
    I wouldn't consider any server doing VPS with only 4GB of RAM.
    Please note we are very security minded and offer a secured /tmp setup, firewall, and a few other things to lock down the VPS before we turn it over to our customer.
    We're slowly moving to Opteron systems as we are finding them to be much faster than the Xeon's for hosting VPS's. We use Opteron 242 processors which equal or beat the Xeon 3.06GHz's (you can search the web for benchmarks if you want to see proof). Since the Opteron's memory controllers are better, have more cache, do better in a multi-processor configuration, have superior process spawning (a big thing), and generate less heat / use less power they have won us over. Also, when Virtuozzo finally releases a 64-bit version them they will perform even better than they do now.
    Do not consider a VPS without a RAID option. Also on that note, find out what kind of RAID it is that they are actually running. For instance, a software RAID or a RAID on an IDE or SATA drive will still be far more inferior to a SCSI RAID.
    We do not allow things like irc on the network, etc.
    (Carlos Note: I stay away from companies that allows IRC.)
    DDOS mitigation devices are extremely expensive
    he only thing that really will mater in your case is the amount of guaranteed RAM for your virtual server. That will determine how many threads of Apache can run simultaneously for example. Do not even look at the MAX RAM numbers, they are irrelevant.
    What is 50% CPU? Do they plan to only put 4 VPSs on a Dual XEON server? There is something seriously wrong there...
    Carlos

  15. #15
    Once again, the comments above came from representatives that I contacted -- I've contacted now over 60 companies!!

    The latest comparison is attached below.

    Regards,
    Attached Files Attached Files
    Carlos

  16. #16
    New comment from hosting company which I found interesting...

    Some VPS providers simply do not limit CPU use on their VPSs. While this makes for some nice benchmarking when servers are tested, we have found that in real-world environments, this does not work well. We don't ever want another customer on your server to be able to affect your performance, so we do indeed place CPU limits on our servers. We'd like to think they are ample and fair, and our hundreds of satisfied customers would hopefully concur. Nonetheless, what this means is that here you are not going to ever be in a position to 'take down' other peoples' servers with your overuse.
    ...all VPS accounts on Dual Xeon 3.06Ghz machines with 15,000rpm Ultra320 RAID-5 arrays The performance gain from these drives over 10,000rpm SCSI or SATA drives alone is significant and unmatched by our competition.
    The following shouldn't be taken too seriously...
    Next bit of information comes from my "VPS vs. Dedicated Compared (Spreadsheet)" http://www.webhostingtalk.com/showth...hreadid=360942

    I was curious as to the average score of VPS vs. Dedicated servers. The range of VPS and dedicated servers is pretty wide, however, this might not be the most scientific way to compare. The question of "real world" performance also should be considered, and perhaps the "scores" listed in that spreadhseet (which comes from a thread on WHT) don't.

    The reason why I bring it up is that, the average VPS score comes out to be 36.4 (dedicated average is 110.9.) With the 36.4 score in mind, I plan to revisit that thread (benchmarking) and see how the VPS plans of the companies I am comparing rate.

    Anyways, here is what I found. I found the average benchmark "score" for each company offering a VPS plan. DefenderHosting and FluidHosting scored over 50. Tektonic and Dinix scored above 20., with ServInt scoring above 10. Note that I only checked out companies who had more than one benchmark done. (i.e. If someone ran a benchmark on SuperDuperHost's VPS plan, and it scored a 40, I didn't include it, since it was only 1 benchmark. Make sense?)

    Not sure why Defender and Fluid had better results. Could be (a) Dual Xeons 3.06 GHz /w RAID that they use (b) less VPS per server (c) better DC/network (d) software (e) a combination of any of these

    Again, don't take this little mini-report to mean that Company A will always be faster than Company B. (And remember, even the best spec/speed is no good if the server has tons of downtime -- ie. support very much matters.)

    I'm very interested to see users of beachcomber.net provide some benchmarks (in the benchmark thread) so that we can see if their Dual Operton servers make a big impact for VPS use.

    Well, I hope this thread is somewhat useful to people looking at VPS plans. I think after my research, and communication with all of these companies, I have about six companies on my short list.

    Cheers!
    Carlos

  17. #17
    Join Date
    Jun 2002
    Location
    PA, USA
    Posts
    5,143
    Some VPS providers simply do not limit CPU use on their VPSs. While this makes for some nice benchmarking when servers are tested, we have found that in real-world environments, this does not work well. We don't ever want another customer on your server to be able to affect your performance, so we do indeed place CPU limits on our servers. We'd like to think they are ample and fair, and our hundreds of satisfied customers would hopefully concur. Nonetheless, what this means is that here you are not going to ever be in a position to 'take down' other peoples' servers with your overuse.
    Am I correct to say that this comment comes from UML based VPS provider?

    UML VPS has larger overhead than VPS and it's the main reason why they score slower. It is true that VPS Virtuozzo has the ability to burst its CPU and memory usage. However, each VPS still has its own guaranteed resources. So at the minimum, each VPS will have its guaranteed resources and when the server is not too busy, why not let some or all VPS to use more than its guaranteed CPU/memory? Seems like the best of all possible worlds.

    The VPS benchmark shows that in general, Virtuozzo based VPS has higher benchmark than other VPS (UML, etc). However, I am not quite sure why ServInt VPS (also Virtuozzo based) score lower. I would like to know this too. In any case, ServInt VPS customers seem to be happy with their VPS/VPS performance and that's all what matters at the end.

    Real world benchmark would be interesting. And I put my bet on Virtuozzo VPS on real world benchmarks ... Somebody should start a thread on this

    I am sure your thread has been of great resources for man, Carlos. Thank you!
    Fluid Hosting, LLC - Enterprise Cloud Infrastructure: Cloud Shared and Reseller, Cloud VPS, and Cloud Hybrid Server

  18. #18
    Thanks, it would be very helpful if aq new thread was created strickly for VPS users. Using the following template:

    Company:
    Name of plan:
    RAM (MB):
    Speed (MHz):
    OS:
    VPS tech (ie UML/Virtuozzo)
    Control Panel:
    Benchmark Score:

    Should I make a new thread to get VPS people to post their benchmarks?
    Carlos

  19. #19
    Join Date
    Jan 2004
    Location
    Pennsylvania
    Posts
    942
    Dave,

    They are using the cpu limit feature of Virtuozzo (vzctl set xx--cpulimit xx%). It puts a hard cap on CPU resources (100% per CPU, so max on a Dual proc system is 200%). We use it a little also hence why our benchmarks are a bit lower. ServInt must have a fairly low setting for this, I imagine 15% of total capacity by the numbers. Ours is tuned quite a bit higher, but it really does prevent a single VPS from hurting anyone else. On some servers we have VPS's sending out newsletters, etc which use a LOT of processing power. By having the limit in place it just slows their VPS down a little bit and no one else is affected. This also means even if there is a heavy usage VPS your CPU will still remain XX% idle for dealing with others, the CPU QoS (cpuunits) doesn't even need to fully go into affect. Much better than having the CPU always pegged at 0% idle.
    Matt Ayres - togglebox.com
    Linux and Windows Cloud Virtual Datacenters powered by Onapp / Xen
    Instant Setup, Instant Scalability, Full Lifecycle Hosting Solutions

    www.togglebox.com

  20. #20
    Join Date
    Dec 2001
    Location
    United Kingdom
    Posts
    77
    I currently use ServInt and wanted to add a few comments...

    Control Panels available are: cPanel, Plesk, DirectAdmin & None

    Also, not sure about being based in the Equinix Asburn VA datacenter - I thought they had their own facilities in McLean, VA?

    You also get 5% discount for prepaying 6 months, and 10% for prepaying 12 months. Also, additional VPS's beyond your first get an additional 10% reduction in price.

    ServInt also have a forum in place, but only for customers.
    Unofficial-Support(.com) | Dedicated To Supporting Your Server

    • Designed, Hosted & Maintained by Zealous Works Ltd • http://www.zealousworks.com/

  21. #21
    misnomer, make sure you are looking at my most recent spreadsheet as I have fixed some issues that you mentioned.

    >ServInt also have a forum in place, but only for customers.
    Yes, I know. I wish companies would at least allow guests to VIEW the forum. I don't ask for posting privilege, but lurking is mighty important IMHO.

    I know some people say the benchmarks are worth considering and some say no. ServInt from my research is top class, if you don't consider the benchmark results, which they seem to rank low in.

    Cheers,
    Carlos

  22. #22
    Join Date
    Dec 2001
    Location
    United Kingdom
    Posts
    77
    I am looking at the most recent attachment you've posted on this thread... http://www.webhostingtalk.com/attach...postid=2856645

    Have you posted a more recent one elsewhere?
    Unofficial-Support(.com) | Dedicated To Supporting Your Server

    • Designed, Hosted & Maintained by Zealous Works Ltd • http://www.zealousworks.com/

  23. #23
    Join Date
    Dec 2001
    Location
    United Kingdom
    Posts
    77
    Also I noticed you've said at the bottom of your sheet "If running Virtuozzo, version to 2.6.1 or higher is a must" - 2.6.1 is only release candidate, so any host offering this is jumping the gun and allowing you to run on pre-release software (non-production). 2.6.0 is the latest release version of Virtuozzo on Linux.
    Unofficial-Support(.com) | Dedicated To Supporting Your Server

    • Designed, Hosted & Maintained by Zealous Works Ltd • http://www.zealousworks.com/

  24. #24
    My apologies. Please see the attachment in this post. Should be dated Feb 15th.

    If running Virtuozzo, version to 2.6.1 or higher is a must" - 2.6.1 is only release candidate, so any host offering this is jumping the gun and allowing you to run on pre-release software (non-production). 2.6.0 is the latest release version of Virtuozzo on Linux.
    Yes, that was a bit of a "test" of the company's knowledge. I know that 2.6.1 hasn't been released yet.

    >However, I am not quite sure why ServInt VPS (also Virtuozzo >based) score lower.
    I would like to know myself. It seems that 98% of their VPS customers are happy from searching in WHT. I wish I could obtain more data on this issue.

    Attached Files Attached Files
    Carlos

  25. #25
    Join Date
    Dec 2001
    Location
    United Kingdom
    Posts
    77
    I've just downloaded the zip again, but the .xls is dated 10th Feb?
    Unofficial-Support(.com) | Dedicated To Supporting Your Server

    • Designed, Hosted & Maintained by Zealous Works Ltd • http://www.zealousworks.com/

Page 1 of 4 1234 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •