Highly Reliable VPS/"Cloud" server

johnk

Hi everyone!

I'm interested in a provider who offers extremely reliable & durable VPS/cloud servers. Here's what I'm looking for:

1 CPU (don't care what model assuming it's not total crap)
1-2 GB RAM
100 GB storage (with room to grow)
100 mbps up/down
400 GB outbound/month
3-5 TB inbound if metered.

Ideally, networked & replicated storage with an AFR < 0.1%, automated compute/instance failover and extremely solid network/power uptime > 99.99%.

Not performance sensitive - as long as its decent, I'll take it.

Budget is $20-40/month.

I've looked at the major public clouds which hit my reliability and price marks (Azure, IBM, Oracle especially), but I'm sure there's a lot of other great hosts out there, I'd love to try too.

Let's hear some offers!

Edit: Sorry, forgot to mention. USA/NA is highly preferred.

ehab

with that budget then stick with the them .. consider adding google.

FAT32

Always think @Clouvider HA VPS is quite premium but sadly never used them before (Also because I missed a deal haha)

https://www.clouvider.com/cloud-vps/

Not as big as other cloud providers but they are decent enough.

ehab

ehab

@FAT32 said:

i think i will get heart attack next day if you paid for one hahahha lol j/k

johnk

@FAT32 said:
Always think @Clouvider HA VPS is quite premium but sadly never used them before (Also because I missed a deal haha)

https://www.clouvider.com/cloud-vps/

Not as big as other cloud providers but they are decent enough.

Yeah, I've looked at them too. Unfortunately, no US location (yet), otherwise they'd fit the bill perfectly.

Forgot to mention that in initial post - I've amended it to add that.

FAT32

@ehab said:
i think i will get heart attack next day if you paid for one hahahha lol j/k

There was an offer quite a while ago that could fit in my impulsive buying range... I asked Dom few times but dont think he wants it happens again

johnk

@ehab said:
with that budget then stick with the them .. consider adding google.

Google is actually the one that's out of budget. LOL

Compute + storage would be ~ $20/m. Bandwidth @ standard rate is $40/m or so alone. I guess it's only a bit over, but there are much better options elsewhere

ehab

@FAT32 said:

time flies so

cybertech

you could check upcloud, they have 100% SLA iirc

FAT32

Can try OVH? https://us.ovhcloud.com/vps/compare/

My SG VPS with them has been running for 3+ years with 100% uptime

(I don't know their SLA though, but based on my experience with them I am confident to use them for production)

johnk

@cybertech said:
you could check upcloud, they have 100% SLA iirc

Thanks! That's on my short list too - we actually had a long chat with them when looking for a primary provider, though that dropped out. Have you had any experiences with them?

johnk

@FAT32 said:
Can try OVH? https://us.ovhcloud.com/vps/compare/

My SG VPS with them has been running for 3+ years with 100% uptime

(I don't know their SLA though, but based on my experience with them I am confident to use them for production)

Thanks! Their pricing is unbeatable for what they offer (as a semi-big provider). Do you happen to know what their storage is backed by? I recall reading ceph somewhere, but that could just be their object storage.

ehab

OVH have great hw,specs,cp etc --- but support is almost 0.

johnk

@ehab said:
OVH have great hw,specs,cp etc --- but support is almost 0.

So I've heard...Assuming no issues on their side, support (or lack of thereof) shouldn't be an issue, though I'd probably rather pay $10-20 more for someone to speak to if something does happen.

ehab

yep, pay $10-20 more - for support.

cybertech

@johnk said:

@cybertech said:
you could check upcloud, they have 100% SLA iirc

Thanks! That's on my short list too - we actually had a long chat with them when looking for a primary provider, though that dropped out. Have you had any experiences with them?

I have credits there so only use them hourly from time to time. couldn't afford long term usage though what I can say is their performance is decent and network is pretty good. limited locations compared to vultr. their maxiops block storage should fit well for expansion.

johnk

@cybertech said:

@johnk said:

@cybertech said:
you could check upcloud, they have 100% SLA iirc

Thanks! That's on my short list too - we actually had a long chat with them when looking for a primary provider, though that dropped out. Have you had any experiences with them?

I have credits there so only use them hourly from time to time. couldn't afford long term usage though what I can say is their performance is decent and network is pretty good. limited locations compared to vultr. their maxiops block storage should fit well for expansion.

I see! Yeah, that (performance, network, storage) was pretty much our experience too with the couple hundred dollars in credit they were kind enough to give us for some in-depth evaluation. We didn't get a chance to evaluate their reliability over a long term though, and I've been hard pressed to find any review about that actually addresses it. I'd spin up a VM and watch it over a few months, but I don't have that kind of timeline unfortunately

skorupion

racknerd (my signature) so far after 5 months? 100% uptime

FAT32

@skorupion said:
racknerd (my signature) so far after 5 months? 100% uptime

NO
NO
NO
NO
NO
NO

NO RackNerd please thanks, I already have enough at OGF + I won't ever define it as highly reliable.

cybertech

@FAT32 said:

@skorupion said:
racknerd (my signature) so far after 5 months? 100% uptime

NO
NO
NO
NO
NO
NO

NO RackNerd please thanks, I already have enough at OGF + I won't ever define it as highly reliable.

i just had a few hours of network downtime days ago, so yeah definitely not 100% uptime.

good support, but not HA for sure.

skorupion

@cybertech said:

@FAT32 said:

@skorupion said:
racknerd (my signature) so far after 5 months? 100% uptime

NO
NO
NO
NO
NO
NO

NO RackNerd please thanks, I already have enough at OGF + I won't ever define it as highly reliable.

i just had a few hours of network downtime days ago, so yeah definitely not 100% uptime.

good support, but not HA for sure.

been solid in AMS for me lolz

cybertech

@skorupion said:

@cybertech said:

@FAT32 said:

@skorupion said:
racknerd (my signature) so far after 5 months? 100% uptime

NO
NO
NO
NO
NO
NO

NO RackNerd please thanks, I already have enough at OGF + I won't ever define it as highly reliable.

i just had a few hours of network downtime days ago, so yeah definitely not 100% uptime.

good support, but not HA for sure.

been solid in AMS for me lolz

racknerd's really good for the price, but OP needs HA, that actually filters out most lowend providers.

otherwise i would have thrown letbox into the list as well with their block storage. 100% uptime for me in 2 years as far as i could see.

johnk

@skorupion said:
racknerd (my signature) so far after 5 months? 100% uptime

I appreciate the response! But, that's probably a bit far from what I'm looking for.

I know the above experience would be far from the only one about network outages/storage failures, though I wouldn't expect much more given the price

tetech

Based on my experience, I'd stick with Azure when such availability is needed. No "low-end" provider I've seen does the same sort of proactive live migrations without downtime.

If you step down a notch you might want to consider LunaNode. They at least let you provision from a block volume, and if I recall correctly they will automatically re-launch on an available node if a block-backed VPS goes down.

Mr_Tom

@FAT32 said: Always think @Clouvider HA VPS is quite premium

Can confirm this is the case, but OP wanted US.

We're on Clouviders network in the US, and Hetrix is showing 99.9998% uptime in NY, but we don't currently have network/replicated storage there.

@FAT32 said: There was an offer quite a while ago that could fit in my impulsive buying range...

I think I got one if it's the offer I'm thinking of.

cybertech

@FAT32 said:

@ehab said:
i think i will get heart attack next day if you paid for one hahahha lol j/k

There was an offer quite a while ago that could fit in my impulsive buying range... I asked Dom few times but dont think he wants it happens again

what kinda offer was that?

FAT32

@cybertech said:
what kinda offer was that?

I don't want to spoil the market so I will just PM you

Hint:

cybertech

@FAT32 said:

@cybertech said:
what kinda offer was that?

I don't want to spoil the market so I will just PM you

Hint:

wow. that was really one off.

Mr_Tom

Oh I don't remember that one?

chimichurri

@johnk said:
I've looked at the major public clouds which hit my reliability and price marks (Azure, IBM, Oracle especially), but I'm sure there's a lot of other great hosts out there, I'd love to try too.

Let's hear some offers!

@johnk said:

@ehab said:
with that budget then stick with the them .. consider adding google.

Google is actually the one that's out of budget. LOL

Compute + storage would be ~ $20/m. Bandwidth @ standard rate is $40/m or so alone. I guess it's only a bit over, but there are much better options elsewhere

I'm afraid Azure too is above your stated budget...
400GB of bandwidth alone is 35 dollars, add to that a crappy B1S (1 vCPU burstable, 10% baseline) instance...

(and you said you need 100 GB of storage too..)

Also, technically, in order to get a promise of 99.99% availability from M$, you'd need at least two of these...
https://azure.microsoft.com/en-us/support/legal/sla/virtual-machines/v1_9/

For all Virtual Machines that have two or more instances deployed across two or more Availability Zones in the same Azure region, we guarantee you will have Virtual Machine Connectivity to at least one instance at least 99.99% of the time.

(otherwise it's just 99.9% at most)

For any Single Instance Virtual Machine using Premium SSD or Ultra Disk for all Operating System Disks and Data Disks, we guarantee you will have Virtual Machine Connectivity of at least 99.9%.

Actually if it's just one server you're looking for (and not let say the same x10), and you'd be OK with either a 1/8 of an Epyc or up to couple of ARM cores / VPS, perhaps just go with Oracle Cloud?
https://www.oracle.com/cloud/free/
Possible downside: Bandwidth might be limited only up to 50Mbps (based on various YABS samples for EPYC instances, example ; unsure about ARM)

@tetech said:
If you step down a notch you might want to consider LunaNode. They at least let you provision from a block volume, and if I recall correctly they will automatically re-launch on an available node if a block-backed VPS goes down.

Alternative (albeit similar) idea, provided Europe would also be tolerable: Hetzner! (has a lot of street cred in these circles and arguably provides what you're looking for)
https://www.hetzner.com/cloud (servers)

Servers with ceph disks store their data on a remote filesystem. Each block is stored on three different servers. They are especially suited for higher availability needs: If the local host hardware fails, we will boot the server on a different machine.

https://docs.hetzner.com/cloud/volumes/faq/ (block storage)

We store every block of data on three different physical servers (triple replication).