Howdy, Stranger!

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


a lesson: learn to buy the right VPS
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.

a lesson: learn to buy the right VPS

This BF I bought dozens of VPS, they are almost useless, due to very poor performance.
So I have got a lesson, don't be cheated by price, what you paid, what you got.
I hope everyone can choose the right VPS for your requirement.
Should not be like me, buying the useless products, wastes money.
I wrote a note for this lesson, if you can read Chinese, could take a look:
https://1drv.ms/b/s!AlsbmZELhAliulE7EwYwJ5Q36pEZ?e=sNJgvZ

Thanks.

«1

Comments

  • Merry Christmas!
    Live and Learn.

    Hope many more will be more judicious in the future sales (in reality, probably not)

    Thanked by 2jpeng skorous
  • dont buy blindly. select the good provider.

    Thanked by 1jpeng
  • @jpeng said:

    I wrote a note for this lesson, if you can read Chinese, could take a look:
    https://1drv.ms/b/s!AlsbmZELhAliulE7EwYwJ5Q36pEZ?e=sNJgvZ

    :) Nice article - Would have beep perfect if it were written in Traditional Chinese :)

    Thanked by 1jpeng
  • tl;dr

    Virmach:
    Very bad I/O
    Very bad cpu
    Bad network (to everywhere..?)

    Pacificrack:
    Very bad I/O
    Bad cpu
    Unstable network (to china..?)

    Naranja:
    All good

    Don't get ripped off, servers that are unusable are a waste of money. Before you buy, check out reviews on hostloc, LET etc. OP is going to idle the servers and let them expire next year.


    Just to add on, there are many good deals with amazing performance for the price. But I'm always somewhat concerned that they will be gone in 1-2 years. It actually is possible to get cheap, fast and good on LET, but always be ready for the business to vanish at any time.

    I continue renewing servers from virmach because the specs on paper seem amazing for the price I pay, which allows me to idle servers on the cheap :smiley:

  • NDTNNDTN Member, Patron Provider, Top Host
  • if io is determined by hardware/hd, it is constant.
    if io is determined by manual adjustment, then io can vary accordingly.

    https://www.lowendtalk.com/discussion/comment/3183372/#Comment_3183372

    Thanked by 1jpeng
  • @NDTN said:
    Rule #1: Buy what you need, not what you want :)

    Easier said than done :p look at my signature >:)

    Thanked by 2NDTN Xiaoming
  • FAT32FAT32 Administrator, Deal Compiler Extraordinaire
    edited December 2020

    @NDTN said:
    Rule #1: Buy what you need, not what you want :)

    I feel offended because this hit me hard

  • @jpeng said:
    This BF I bought dozens of VPS, they are almost useless, due to very poor performance.
    So I have got a lesson, don't be cheated by price, what you paid, what you got.
    I hope everyone can choose the right VPS for your requirement.
    Should not be like me, buying the useless products, wastes money.
    I wrote a note for this lesson, if you can read Chinese, could take a look:
    https://1drv.ms/b/s!AlsbmZELhAliulE7EwYwJ5Q36pEZ?e=sNJgvZ

    Thanks.

    This is an English site. Please keep that in mind.

    Did you just benchmark or did you actually try your intended app on it and it failed?

    Benchmarks aren't all that and are just one thing to use to measure performance. It's really to help confirm whether you're having problems. For me, the best torrenting server I have has lowest GB5 score out of over a dozen. It doesn't have the worst network (it's great), but despite several other servers with 10Gbps speeds being excellent and multiple times faster, it doesn't matter if it's not actually faster in practice.

  • @jpeng said:
    This BF I bought dozens of VPS, they are almost useless, due to very poor performance.
    So I have got a lesson, don't be cheated by price, what you paid, what you got.
    I hope everyone can choose the right VPS for your requirement.
    Should not be like me, buying the useless products, wastes money.
    I wrote a note for this lesson, if you can read Chinese, could take a look:
    https://1drv.ms/b/s!AlsbmZELhAliulE7EwYwJ5Q36pEZ?e=sNJgvZ

    Thanks.

    Share to me please.

  • This is why people should do two things before buying a VPS:

    1. Read comments here extensively and do not buy just because it is "cheap".
    2. Read point 1 again.
    Thanked by 1jsg
  • Liteserver = everything perfect

    Thanked by 1k4zz
  • The good thing is you realised and learned from it, so many other people can benefit from kind providers here.

  • not sure what you were expecting out of a virmach BF special.

  • @poisson said:
    This is why people should do two things before buying a VPS:

    1. Read comments here extensively and do not buy just because it is "cheap".
    2. Read point 1 again.

    pls share with us your purchase story this season.

  • I am not so familiar with policy here. can I resell my unused VMs to other members on LET by making a post? thanks

  • FAT32FAT32 Administrator, Deal Compiler Extraordinaire

    @jpeng said:
    I am not so familiar with policy here. can I resell my unused VMs to other members on LET by making a post? thanks

    That's usually fine as long as it is not a bug promo / free stuff / resell at a significant higher price than the actual price.

    Thanked by 1jpeng
  • b@n4na detected :blush:

  • raindog308raindog308 Administrator, Veteran

    @smallbibi said: Virmach:

    Very bad I/O Very bad cpu Bad network (to everywhere..?)

    Here comes @VirMach 's 5,000-word response...

  • When buying a VPS, get the details of the host server if possible and check if they are being oversold. Do not go for a HDD instead of SSD thinking you are getting more space at a cheaper price as The performance will suffer even if you have a good CPU and plenty of RAM. Many hosting providers oversell disk space and so it is also recommended to ask the host how much disk i/o they guarantee.

    Thanked by 2jpeng JasonM
  • forcesforces Member
    edited December 2020

    @CjayVPS said:
    Liteserver = everything perfect

    words of gold, always remember this. best advice you can have now.

    "今年黑五期间,我空虚寂寞加手贱"
    feeling lonely, emptiness and being simply a moron during BF
    bro, its now christmas and new year and CNY to come
    be strong bro @jpeng , be strong
    Btw, reviewing @naranjatech so highly soon MJJ will swarm to conquer naranja tech...
    i do hope they will tighten their identity check from now on...

    Thanked by 2jpeng k4zz
  • elliotcelliotc Member
    edited December 2020

    Another lesson: people never learn from their past. :)

    Thanked by 1seriesn
  • jsgjsg Member, Resident Benchmarker

    @jpeng

    Well, at least in part the bad performance of your new BF VPS might be due to almost everyone "stress testing" and benchmarking their new VPS.

    Thanked by 1jpeng
  • If you like the provider and think they do a decent job, participate in the forum and generally don't suck - buy a yearly from them to support them and have a vps to tinker with. However, if its for production- go to one of the hourly providers or monthly providers where your risks are mitigated - and it allows you to jump ship on the early signs of stupidity.

    Buying multiple VPS from a provider on here may not be wise, even if they are the golden boy for the site owner or one of CC's favorite. Things change

    Thanked by 2seriesn jpeng
  • I have finished reading your article. The performance of the Dutch VPS you praised is very similar to a Dutch VPS-Clouveo I bought some time ago. It is about $6.7, which is my first prize in Black Friday.

  • @waynechris said:
    I have finished reading your article. The performance of the Dutch VPS you praised is very similar to a Dutch VPS-Clouveo I bought some time ago. It is about $6.7, which is my first prize in Black Friday.

    Looks good.

  • @forces said:

    @CjayVPS said:
    Liteserver = everything perfect

    word of gold, always remember this. best advice you can have now.

    "今年黑五期间,我空虚寂寞加手贱"
    feeling lonely, emptiness and being simply a moron during BF
    bro, its now christmas and new year and CNY to come
    be strong bro @jpeng , be strong
    Btw, reviewing @naranjatech so highly soon MJJ will swarm to conquer naranja tech...
    i do hope they will tighten their identity check from now on...

    Thanks for your warm words.

  • darkimmortaldarkimmortal Member
    edited December 2020

    @CjayVPS said:
    Liteserver = everything perfect

    They are just a few days away from running OpenVZ 6 in 2021

    Following the spirit of this thread, another hidden factor of VPS purchases is node maintenance. Providers need to strike a good balance between uptime/‘if it ain’t broke don’t fix it’ and security. Even with KVM I don’t want the node to be full of holes

  • @smallbibi said:
    tl;dr

    Virmach:
    Very bad I/O
    Very bad cpu
    Bad network (to everywhere..?)

    I disagree with disk & cpu being very bad and network being bad. Network and disk performance is pretty good, also CPU is a little weak, but acceptable for the price range..

    This is my BF2020 in LA for $4.95/yr.

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2020-12-27                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Sun 27 Dec 2020 06:33:29 AM EST
    
    Basic System Information:
    ---------------------------------
    Processor  : Intel Xeon E312xx (Sandy Bridge, IBRS update)
    CPU cores  : 1 @ 2499.998 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 987.5 MiB
    Swap       : 256.0 MiB
    Disk       : 14.5 GiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 27.70 MB/s    (6.9k) | 296.39 MB/s   (4.6k)
    Write      | 27.71 MB/s    (6.9k) | 297.95 MB/s   (4.6k)
    Total      | 55.42 MB/s   (13.8k) | 594.34 MB/s   (9.2k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 573.68 MB/s   (1.1k) | 639.43 MB/s    (624)
    Write      | 604.16 MB/s   (1.1k) | 682.02 MB/s    (666)
    Total      | 1.17 GB/s     (2.3k) | 1.32 GB/s     (1.2k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 1.08 Gbits/sec  | 777 Mbits/sec  
    Online.net      | Paris, FR (10G)           | 893 Mbits/sec   | 677 Mbits/sec  
    WorldStream     | The Netherlands (10G)     | 954 Mbits/sec   | 661 Mbits/sec  
    Biznet          | Jakarta, Indonesia (1G)   | 810 Mbits/sec   | 32.3 Mbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 1.72 Gbits/sec  | busy           
    Velocity Online | Tallahassee, FL, US (10G) | 2.04 Gbits/sec  | 1.03 Gbits/sec 
    Clouvider       | Los Angeles, CA, US (10G) | 4.44 Gbits/sec  | 3.34 Gbits/sec 
    Iveloz Telecom  | Sao Paulo, BR (2G)        | 453 Mbits/sec   | 440 Mbits/sec 
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 338                           
    Multi Core      | 339                           
    Full Test       | https://browser.geekbench.com/v5/cpu/5554040
    

    and this is BF2018 in Amsterdam for $10/yr.

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2020-12-27                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Sun 27 Dec 2020 06:38:47 PM WIB
    
    Basic System Information:
    ---------------------------------
    Processor  : Intel(R) Xeon(R) CPU E5-2687W v2 @ 3.40GHz
    CPU cores  : 2 @ 3399.998 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 1.9 GiB
    Swap       : 1.2 GiB
    Disk       : 14.5 GiB
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 57.88 MB/s   (14.4k) | 507.06 MB/s   (7.9k)
    Write      | 57.97 MB/s   (14.4k) | 509.73 MB/s   (7.9k)
    Total      | 115.86 MB/s  (28.9k) | 1.01 GB/s    (15.8k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 727.52 MB/s   (1.4k) | 724.06 MB/s    (707)
    Write      | 766.17 MB/s   (1.4k) | 772.29 MB/s    (754)
    Total      | 1.49 GB/s     (2.9k) | 1.49 GB/s     (1.4k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed     
                    |                           |                 |                
    Clouvider       | London, UK (10G)          | 1.27 Gbits/sec  | busy           
    Online.net      | Paris, FR (10G)           | 1000 Mbits/sec  | 404 Mbits/sec  
    WorldStream     | The Netherlands (10G)     | busy            | 791 Mbits/sec  
    Biznet          | Jakarta, Indonesia (1G)   | 893 Mbits/sec   | 91.4 Mbits/sec 
    Clouvider       | NYC, NY, US (10G)         | 1.34 Gbits/sec  | busy           
    Velocity Online | Tallahassee, FL, US (10G) | 1.64 Gbits/sec  | 225 Mbits/sec  
    Clouvider       | Los Angeles, CA, US (10G) | busy            | 269 Mbits/sec  
    Iveloz Telecom  | Sao Paulo, BR (2G)        | 602 Mbits/sec   | 133 Mbits/sec  
    
    Geekbench 5 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 468                           
    Multi Core      | 777                           
    Full Test       | https://browser.geekbench.com/v5/cpu/5554069
    
    Thanked by 2skorous coreflux
  • @chocolateshirt said: I disagree with disk & cpu being very bad and network being bad. Network and disk performance is pretty good, also CPU is a little weak, but acceptable for the price range..

    I just wanna point out that I was translating what op had in his report. It doesn't reflect my opinions, and I agree it is acceptable for what you pay.

    Thanked by 1chocolateshirt
Sign In or Register to comment.