Howdy, Stranger!

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


Looking for a Production 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.

Looking for a Production VPS

DumbledoreDumbledore Member
edited August 2017 in Requests
CPU: 2 Dedicated Cores
RAM: 6GB or More
Storage: 120GB(SSD) or More in RAID 10
DC :  A premium datacenter preferred. 
Loc : US / UK

I already have Incero and netcup.

Comments

  • PUSHR_VictorPUSHR_Victor Member, Host Rep

    Not sure if cores are dedicated with RamNode, but I've only heard good things about them as a provider. Have not used them personally.

  • Get a Slice @Francisco

    Thanked by 1Francisco
  • For a production use, can we rely on slice?

  • sonicsonic Veteran

    @ErawanArifNugroho said:
    For a production use, can we rely on slice?

    Why not?

  • AuroraZ said: Get a Slice @Francisco

    sonic said: Get a slice!

    I ate my share of Slice :P. I need a different provider for this.

  • @Dumbledore said:
    CPU: 2 Dedicated Cores
    RAM: 6GB or More
    Storage: 120GB(SSD) or More in RAID 10
    DC : A premium datacenter preferred.
    Loc : US / UK

    I already have Incero and netcup.

    For a production use, can we rely on slice?

    Of course not!

    A single slice for production is not production. Nobody hosts production stuff on a single server/slice, they have backs ups on 1 or more regions, or at least a back up setup on standby.

    If your single slice or server goes down, the provider is not to blame but yourself because you think production means that they won't ever go down. In fact they do, and if you want a host to provide you with 4-5 nines, then be ready to pay in the $$$$-$$$$$ a month for that kind of commitment.

    Do you have a budget of $$$$-$$$$$ for actual production use?

  • @IAlwaysBeCoding said:

    @Dumbledore said:
    CPU: 2 Dedicated Cores
    RAM: 6GB or More
    Storage: 120GB(SSD) or More in RAID 10
    DC : A premium datacenter preferred.
    Loc : US / UK

    I already have Incero and netcup.

    For a production use, can we rely on slice?

    Of course not!

    A single slice for production is not production. Nobody hosts production stuff on a single server/slice, they have backs ups on 1 or more regions, or at least a back up setup on standby.

    If your single slice or server goes down, the provider is not to blame but yourself because you think production means that they won't ever go down. In fact they do, and if you want a host to provide you with 4-5 nines, then be ready to pay in the $$$$-$$$$$ a month for that kind of commitment.

    Do you have a budget of $$$$-$$$$$ for actual production use?

    I don't think you need that big budget for 99.99. He's asking if he can. Yes

  • IAlwaysBeCoding said: Of course not!

    A single slice for production is not production. Nobody hosts production stuff on a single server/slice, they have backs ups on 1 or more regions, or at least a back up setup on standby.

    If your single slice or server goes down, the provider is not to blame but yourself because you think production means that they won't ever go down. In fact they do, and if you want a host to provide you with 4-5 nines, then be ready to pay in the $$$$-$$$$$ a month for that kind of commitment.

    Do you have a budget of $$$$-$$$$$ for actual production use?

    I think you are replying to the wrong person.

    I have contingencies in place for my "PRODUCTION" deployments. I normally use Softlayer with HA and multiple offsite backups. Here the client is on a budget so I'm looking for an alternative. There are many providers here who have nodes at Softlayer / Equinix. What I meant is I need a decent provider who won't go down every 48 hours.

    Currently, I'm looking at @Clouvider

  • brueggusbrueggus Member, IPv6 Advocate

    Dumbledore said:
    Currently, I'm looking at @Clouvider

    You won't be disappointed.

    Thanked by 2Clouvider Ympker
  • @Dumbledore said:

    IAlwaysBeCoding said: Of course not!

    A single slice for production is not production. Nobody hosts production stuff on a single server/slice, they have backs ups on 1 or more regions, or at least a back up setup on standby.

    If your single slice or server goes down, the provider is not to blame but yourself because you think production means that they won't ever go down. In fact they do, and if you want a host to provide you with 4-5 nines, then be ready to pay in the $$$$-$$$$$ a month for that kind of commitment.

    Do you have a budget of $$$$-$$$$$ for actual production use?

    I think you are replying to the wrong person.

    I have contingencies in place for my "PRODUCTION" deployments. I normally use Softlayer with HA and multiple offsite backups. Here the client is on a budget so I'm looking for an alternative. There are many providers here who have nodes at Softlayer / Equinix. What I meant is I need a decent provider who won't go down every 48 hours.

    Currently, I'm looking at @Clouvider

    Oh okay, that makes sense now, sorry about that. But why would you look for a production VPS. In my view, production VPS is very close to an oxymoron. I believe you should look for a production budget dedicated server from here.

  • IAlwaysBeCoding said: I believe you should look for a production budget dedicated server from here.

    I know. I just don't want to deal with SolusVM ordeal. I have talked with a couple of providers here for a managed solution. But it got delayed.

    I haven't heard about this host. Can someone share some reviews?

  • They're a pretty professional outfit. Their new KVM lineup is fully refreshed to compete with 2017 LEB prices, with the bonus of dedicated resources.

    Amsterdam should be perfect of you're in Europe.

    Thanked by 2hosthatch bersy
  • AnthonySmithAnthonySmith Member, Patron Provider

    I don't think there are many hosts using raid 10 SSD within LET prices, at least not ones that are telling the truth about it anyway :)

    If you are interested in SAS Raid 10 + SSD cachecade I can give you a price.

  • hosthatchhosthatch Patron Provider, Top Host, Veteran

    @AnthonySmith said:
    I don't think there are many hosts using raid 10 SSD within LET prices, at least not ones that are telling the truth about it anyway :)

    Some really are though :) We take it one step further by putting in 2x 180 GB Intel SSDs in RAID1 for the OS/templates/ISOs, etc.

    [abdullah@nvme1]~% cat /proc/mdstat
    Personalities : [raid1] [raid10]
    md124 : active raid10 nvme1n1p1[1] nvme0n1p1[0] nvme3n1p1[3] nvme2n1p1[2]
          1953260544 blocks super 1.2 512K chunks 2 near-copies [4/4] [UUUU]
          bitmap: 7/15 pages [28KB], 65536KB chunk
    
    md125 : active raid1 sdb3[1] sda3[0]
          141581312 blocks super 1.2 [2/2] [UU]
          bitmap: 1/2 pages [4KB], 65536KB chunk
    
    md126 : active raid1 sdb2[1] sda2[0]
          525248 blocks super 1.0 [2/2] [UU]
          bitmap: 0/1 pages [0KB], 65536KB chunk
    
    md127 : active raid1 sdb1[1] sda1[0]
          33554432 blocks super 1.2 [2/2] [UU]
    
    [root@nvme1 ~]# fdisk -l
    
    Disk /dev/nvme1n1: 1000.2 GB, 1000204886016 bytes, 1953525168 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x0006654d
    
            Device Boot      Start         End      Blocks   Id  System
    /dev/nvme1n1p1            2048  1953525167   976761560   fd  Linux raid autodetect
    
    Disk /dev/nvme0n1: 1000.2 GB, 1000204886016 bytes, 1953525168 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x00091d6f
    
            Device Boot      Start         End      Blocks   Id  System
    /dev/nvme0n1p1            2048  1953525167   976761560   fd  Linux raid autodetect
    
    Disk /dev/nvme3n1: 1000.2 GB, 1000204886016 bytes, 1953525168 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x000addf9
    
            Device Boot      Start         End      Blocks   Id  System
    /dev/nvme3n1p1            2048  1953525167   976761560   fd  Linux raid autodetect
    
    Disk /dev/nvme2n1: 1000.2 GB, 1000204886016 bytes, 1953525168 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x000e8c89
    
            Device Boot      Start         End      Blocks   Id  System
    /dev/nvme2n1p1            2048  1953525167   976761560   fd  Linux raid autodetect
    
    Disk /dev/sda: 180.0 GB, 180045766656 bytes, 351651888 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x000481c8
    
       Device Boot      Start         End      Blocks   Id  System
    /dev/sda1            2048    67176447    33587200   fd  Linux raid autodetect
    /dev/sda2   *    67176448    68227071      525312   fd  Linux raid autodetect
    /dev/sda3        68227072   351651839   141712384   fd  Linux raid autodetect
    
    Disk /dev/sdb: 180.0 GB, 180045766656 bytes, 351651888 sectors
    Units = sectors of 1 * 512 = 512 bytes
    Sector size (logical/physical): 512 bytes / 512 bytes
    I/O size (minimum/optimal): 512 bytes / 512 bytes
    Disk label type: dos
    Disk identifier: 0x0005ca41
    
       Device Boot      Start         End      Blocks   Id  System
    /dev/sdb1            2048    67176447    33587200   fd  Linux raid autodetect
    /dev/sdb2   *    67176448    68227071      525312   fd  Linux raid autodetect
    /dev/sdb3        68227072   351651839   141712384   fd  Linux raid autodetect
    
    Thanked by 1Dumbledore
  • AnthonySmithAnthonySmith Member, Patron Provider

    Abdullah said: Some really are though :)

    And the only KVM package you have that meets the OP's requirements is $40 p/month.

  • hosthatchhosthatch Patron Provider, Top Host, Veteran

    @AnthonySmith said:
    And the only KVM package you have that meets the OP's requirements is $40 p/month.

    I couldn't find the OP's budget in the thread.

    Either way though, we can combine plans, so 8 GB + 2 GB plans will give the OP 125 GB SSD for $25 per month.

    Thanked by 1bersy
  • AnthonySmithAnthonySmith Member, Patron Provider

    Abdullah said: I couldn't find the OP's budget in the thread.

    Yep, don't get me wrong, it is awesome that you are doing NVMe raid 10 (not sure why you have not set up your mdadm.conf properly or hurting performance with bitmaps) but you responded to my point:

    AnthonySmith said: I don't think there are many hosts using raid 10 SSD within LET prices

    And I stand by it, you're not doing that even remotely close to LET prices, you are not even a mile away from softlayer prices.

  • hosthatchhosthatch Patron Provider, Top Host, Veteran
    edited August 2017

    AnthonySmith said: hurting performance with bitmaps

    You're right - disabling it might make it better than hosthatch.com/benchmarks - so definitely should look into it. Always happy to take some positive advice.

    AnthonySmith said: I don't think there are many hosts using raid 10 SSD within LET prices

    And I stand by it, you're not doing that even remotely close to LET prices, you are not even a mile away from softlayer prices.

    I keep missing the part where LET prices were mentioned in this thread. I highly doubt he was looking for $7/mo for these specs.

    (PS - we have 2 plans in the LET range - both also have RAID10)

  • @Abdullah said:

    AnthonySmith said: hurting performance with bitmaps

    You're right - disabling it might make it better than hosthatch.com/benchmarks - so definitely should look into it. Always happy to take some positive advice.

    AnthonySmith said: I don't think there are many hosts using raid 10 SSD within LET prices

    And I stand by it, you're not doing that even remotely close to LET prices, you are not even a mile away from softlayer prices.

    I keep missing the part where LET prices were mentioned in this thread. I highly doubt he was looking for $7/mo for these specs.

    (PS - we have 2 plans in the LET range - both also have RAID10)

    He mentions that he regularly uses " I normally use Softlayer with HA and multiple offsite backups. " for his normal project. Judging from that, he is spending a lot of money on his regular projects. So I assume someone like him wouldn't be dealing with clients spending $xxxx-$xxxxx a month and then looking for something for his other "client" that wants to spend $x a month.

    I can't even tell what "budget" for someone like him means, since he is saying he does setups that obviously are going to cost $xxxx-$xxxxx a month.

  • AnthonySmithAnthonySmith Member, Patron Provider
    edited August 2017

    Fair enough, sorry to derail :) I was just making a point that LET price 'ranges' don't allow for raid 10 ssd and profit :)

    Thanked by 2Clouvider hosthatch
  • DumbledoreDumbledore Member
    edited August 2017

    Abdullah said: Either way though, we can combine plans, so 8 GB + 2 GB plans will give the OP 125 GB SSD for $25 per month.

    Yes. I have talked to Abdullah about this. He is willing to do a custom plan.

    AnthonySmith said: If you are interested in SAS Raid 10 + SSD cachecade I can give you a price.

    I was about to contact you. @Clouvider referred you. Could you PM the offer you have for your London location.

    IAlwaysBeCoding said: he is spending a lot of money on his regular projects

    I already have a price lock with Softlayer for their bare-metal servers. They even uphold their old 20TB bandwidth for all our new orders. But for the cloud, they don't do any significant discounts.

    Thanked by 2Clouvider hosthatch
  • AnthonySmithAnthonySmith Member, Patron Provider

    Dumbledore said: Could you PM the offer you have for your London location.

    Done.

    Thanked by 2Dumbledore Clouvider
  • For a production VPS on a budget consider SSDNodes. Performance/Value ratio is excellent. They use raid 10 ssd.

    Thanked by 1Dumbledore
  • @Mark_O_Polo said:
    For a production VPS on a budget consider SSDNodes. Performance/Value ratio is excellent. They use raid 10 ssd.

    Running production servers on a OpenVZ platform is not highly recommended.

  • MasonRMasonR Community Contributor

    @IAlwaysBeCoding said:

    @Mark_O_Polo said:
    For a production VPS on a budget consider SSDNodes. Performance/Value ratio is excellent. They use raid 10 ssd.

    Running production servers on a OpenVZ platform is not highly recommended.

    They're launching KVM nodes shortly: https://www.ssdnodes.com/kvm-preview/

  • @MasonR said:

    @IAlwaysBeCoding said:

    @Mark_O_Polo said:
    For a production VPS on a budget consider SSDNodes. Performance/Value ratio is excellent. They use raid 10 ssd.

    Running production servers on a OpenVZ platform is not highly recommended.

    They're launching KVM nodes shortly: https://www.ssdnodes.com/kvm-preview/

    Oh that's fucking cool! I originally wanted to try them out as I've heard of them and they have a really cool price. Very tempting, thanks for letting me know about their upcoming KVM.

    Thanked by 1MasonR
  • MasonRMasonR Community Contributor

    @IAlwaysBeCoding said: Oh that's fucking cool! I originally wanted to try them out as I've heard of them and they have a really cool price. Very tempting, thanks for letting me know about their upcoming KVM.

    Sure thing! I had a yearly OVZ VPS with them and their support/hardware/network was top notch. Had to let it go as I've since moved to dedi hoarding :P

  • Mark_O_PoloMark_O_Polo Member
    edited August 2017

    @IAlwaysBeCoding said:
    Running production servers on a OpenVZ platform is not highly recommended.

    True only to some extent (i believe it's overhyped), it really maters how sensitive your data is and your risk assessment. One can make the statement that all virtualizations are subject to exploitation (especially by the sysadmin).

    I wouldn't make a blanket statement against OpenVZ. That's ridiculous.

    We live in a world of insecurity (including dedi's with remote access controllers), but that needs to be a separate thread...

Sign In or Register to comment.