Howdy, Stranger!

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


What VPS/Dedi provider do you use in production?
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.

What VPS/Dedi provider do you use in production?

I am curious what hosting companies LET members are using when they are hosting critical systems that need 100% uptime. Either for clients or themselves.

Personally I wouldn't use most low-end companies listed on the site for such task. What is your opinion?

«1

Comments

  • Just because a provider is listed around LEB/LET doesn't necessarily mean that they don't have other more suitable products that aren't hosted in a different type of environment.

    In our case though, we have different brands for different business focus all owned by one company. VortexNode (which I represent) being our budget brand.

    If you look at UK2 Group as an example. They have 100TB.com, UK2.net, VPS.net and VI.net (plus more). VI.net is their premium brand and targeted at a different market than their others. Most larger companies will have a similar structure.

  • OVH. Not exactly high-end but I know they have lots of experience of making sure the hardware and network is as it should be.

    I use a lot of shared and VPS hosts too. Everything is backed up and if one disappears, things can be easily replaced elsewhere.

  • I also use OVH dedicated servers in production in some cases. Not sure if I trust their VPS yet.

  • I use various providers for sites, and have load-balancers in place, so far my system's had no downtime in the past year.

    Thanked by 1Frecyboy
  • You know what, 100% uptime is a struggle, no matter who you host with. Top-tier hosting companies don't manage it, I've seen some horrendous outages from the big boys over the last few years.

    For anything that needs 100% uptime, you build resiliency into your configurations. If you have that, most of the top 10 on LET would be fine.

    Thanked by 2GCat Falzo
  • resiliency is obvious, but its still best when it never has to come into play.
    I am just wondering what providers people trust for very serious money-making sites/services, where downtime is unacceptable.

  • jimaek said: resiliency is obvious, but its still best when it never has to come into play. I am just wondering what providers people trust for very serious money-making sites/services, where downtime is unacceptable.

    There is always downtime, that's the point. No-one has zero downtime and if they have so far, something just round the corner. You cannot avoid it.

    The only provider I would trust not to fuck something up and cause unnecessary downtime is Rackspace, they're the only provider I've used with zero unscheduled downtime as a result of their own actions.

    Thanked by 1GCat
  • TheLinuxBugTheLinuxBug Member
    edited September 2016

    Honestly, I would be a little bit hesitant to post any company here and make crazy claims about 'uptime', etc. This seems to me an open invitation to get a DDOS based on history of just hosts posting their test IPs.

    No host is perfect, if they haven't had any downtime in their history, so far, they have been very lucky.

    It may not be the best idea to come and spam hosts and make wild claims here to then later find they have been the target of massive DDOS.

    Just my 2 cents.

    Cheers!

  • I am sorry but if a provider will go down because someone mentioned him on some forum I dont think thats a provider anyone should be using at all. I didnt ask for IPs, just for company names.

  • Ovh, Ikoula, leaseweb and i3d. All dedicated. Using some VMs as well from some budget companies. Mostly I run a test setup for a year or so. If they prove to be stable I'll use them and make sure there's some failover in place.

    Thanked by 2jimaek Ikoula
  • main market europe: hosteurope, hetzner, ovh, leaseweb. just to name a few.

    as nekki said, 100% is a lie. if you want to achieve that it's your own reponsibility to setup your own HA/cloud/cluster/whatever accordingly. at least using more than just one provider with proper fallback mechanisms.

    reminds me of Audi attending the 24h of LeMans from 1999. they simply did not aim for an hopefully unbreakable car or engine but on how to have enough spare parts and get them in and out as quick as possible.

  • jimaek said: I am sorry but if a provider will go down because someone mentioned him on some forum I dont think thats a provider anyone should be using at all.

    LET providers only have limited capacity to tank denial of service attacks, and many of them have reactive measures in place rather than proactive. If someone decides to take a run at a provider due to being mentioned here and they go down briefly while getting countermeasures in play, it's hardly their fault.

  • iwaswrongonceiwaswrongonce Member
    edited September 2016

    OVH, AWS, WSI and DO

    Backups stored on OVH, Online.net, SpeedyKVM and B2

  • Hertzner, Kimsufi, Vultr and DO.

  • MikePTMikePT Moderator, Patron Provider, Veteran

    So, let's see:

    OVH, Trabia, RamNode, Dedistation, Infinity's services, Inceptionhosting, VPSDime (and Backupsy, Winity), Rackspace for email, Amazon for email delivery in large-scale.
    Currently considering other providers to host a new brand we'll launch soon.

  • OVH, VULTR, and Leaseweb. I keep a ton of backups so if shit hits the fan I can move stuff around.

  • +1 Leaseweb. They keep the price low without sacrifices support.

  • OVH and LiquidWeb

  • BrianHarrisonBrianHarrison Member, Patron Provider
    edited September 2016

    I see a lot of OVH mentions here. I personally would never consider OVH if I realistically needed five 9s (or better) of uptime.

    One of our clients runs mission critical healthcare web services that have a real-time impact on the quality of care in hospitals across the USA. We host their equipment in Internap datacenters. 24/7/365 remote hands, redundant everything, offsite backups, full HIPAA compliance with the annual SOC 2 Type 2 audits to back it up. The other providers listed above don't come close to what Internap offers.

    A real commitment to high-availability costs a lot of money.

    Thanked by 2BeardyUnixGuy Amitz
  • I'm glad you spoke up about your experiences there, BrianHarrison. I've never personally required something as mission critical as what you describe so couldn't recommend anything based on experience. I think my upper tier of requirement is colo in South England with people I know nearby able to visit the datacenter...

    It'd seem like the simple bottom line is more redundancy and more people on standby to correct anything that goes wrong. How much were you paying Internap and do you have anyone comparable that you'd feel are in a similar league?

    Would be interesting to know the associated price tags also, just to illustrate the exponential premium for those extra 9's in uptime.

  • RXWatcherRXWatcher Member
    edited September 2016

    @BrianHarrison said:

    One of our clients runs mission critical healthcare web services that have a real-time impact on the quality of care in hospitals across the USA. We host their equipment in Internap datacenters. 24/7/365 remote hands, redundant everything, offsite backups, full HIPAA compliance with the annual SOC 2 Type 2 audits to back it up. The other providers listed above don't come close to what Internap offers.

    I work at one of the largest healthcare IT companies out there. Why don't they use Azure? Thats where we're hosting our cloud apps..we have some iaas and paas apps in there, MS does a really impressive job.

  • BrianHarrisonBrianHarrison Member, Patron Provider
    edited September 2016

    @RXWatcher said:

    @BrianHarrison said:

    One of our clients runs mission critical healthcare web services that have a real-time impact on the quality of care in hospitals across the USA. We host their equipment in Internap datacenters. 24/7/365 remote hands, redundant everything, offsite backups, full HIPAA compliance with the annual SOC 2 Type 2 audits to back it up. The other providers listed above don't come close to what Internap offers.

    I work at one of the largest healthcare IT companies out there. Why don't they use Azure? Thats where we're hosting our cloud apps..we have some iaas and paas apps in there, MS does a really impressive job.

    Does your company use the apps to store PHI on Azure services? Labs, vitals, notes, visits, etc with patient identifiers? If so, then you'd need to use Azure's HIPAA compliant services. Your organization signed a BAA with Microsoft and no doubt spent a great deal to use of Microsoft's HIPAA compliant Azure ;) We did things a bit more cost effectively for our client.

    As far as I'm aware, Azure was not HIPAA compliant until recently. The standards for HIPAA compliance are extremely high. For the major geographically dispersed cloud providers, HIPAA compliance is very difficult to implement.

    Expensive comprehensive auditing (e.g., SOC 2 Type 2) is required annually for the entire org -- not many small organizations can afford them every year and many large organizations balk at the cost.

    The hosting provider also needs to enter into a BAA with the client (e.g., a hospital) -- most of the major cloud providers can't take on the liability that a BAA entails.

    If you work with patient data from a hospital, then that hospital's lawyers will likely have their own set of standards for you to comply with.

    Here's the brief primer on Azure's HIPAA compliant services: https://www.microsoft.com/en-us/TrustCenter/Compliance/HIPAA Even from the marketing materials you can tell it's a complex and expensive undertaking.

  • NekkiNekki Veteran
    edited September 2016

    @BrianHarrison said:

    The hosting provider also needs to enter into a BAA with the client (e.g., a hospital) -- most of the major cloud providers can't take on the liability that a BAA entails.

    Surely that would have to be a near-unlimited level of liability? Could you imagine the repercussions of something like a data breach for a healthcare provider?

  • For me.

    Front end - 7 VMs globally diffeeent networks, backbones mostly KVM

    Backend/storage: ObjStore + Time4VPS (as failover) mounted locally. All KVMs have required files cached locally

    Thanked by 1Frecyboy
  • mine OVH (vps), DO, Ramnode and consider using Leaseweb too.

  • dedi - OVH, Kimsufi, Joe's DC, VPS- Leaseweb, DO

  • Codero :)

  • BrianHarrisonBrianHarrison Member, Patron Provider

    @Nekki said:

    @BrianHarrison said:

    The hosting provider also needs to enter into a BAA with the client (e.g., a hospital) -- most of the major cloud providers can't take on the liability that a BAA entails.

    Surely that would have to be a near-unlimited level of liability? Could you imagine the repercussions of something like a data breach for a healthcare provider?

    Depends on how bad their screw up is I suppose. But yeah, when you're dealing with PHI the consequences of screwing up can be pretty big.

  • @Dazzle said:
    +1 Leaseweb. They keep the price low without sacrifices support.

    Ummm...read their SLA and get back to us on that. Also cross your fingers you never have a hardware failure on their standard (up to 24hrs response and replacement time) plan.

  • sman said: Ummm...read their SLA and get back to us on that. Also cross your fingers you never have a hardware failure on their standard (up to 24hrs response and replacement time) plan.

    I've called them (by phone) a number of times when I had an issue. Was usually resolved within an hour. Clearly they will not sit down for 24 hours and help you then! (Could still happen of course, and if it happens it would hurt apps in production...)

Sign In or Register to comment.