5 ms·

I see a lot of mentions in the comments about just using the basic storage/networking/compute from AWS/AZ/GCP--if that's all you're using, you should really con

I see a lot of mentions in the comments about just using the basic storage/networking/compute from AWS/AZ/GCP--if that's all you're using, you should really consider other providers. Linode, Digital Ocean, and Vultr will be far more competitive and offer faster machines, cheaper, and with better bandwidth pricing. The point of using AWS/AZ/GCP is to leverage their managed service portfolio and be locked in. If you aren't doing that, there are better companies that want your business and will treat you much better.

ithrow 6y ago
IME, the network of AWS is much better than that of DO, Linode.
- dilyevsky 6y ago
  How so?
  - ithrow 6y ago
    Less hiccups and downtime. It's faster and with better latency to other third party services. Superior internal control. Ex: In, linode a private IP address gives EVERYONE on the same data center access to your Linode server. Also, last time I used them they didn't have a Firewall.
    - dilyevsky 6y ago
      Right linode is basically old school dedicated servers afaik but DO should be in different class
      - dijit 6y ago
        AWS networking isn't _great_ but it's decidedly better than DO (which is actually the worst of those listed based on my own TCP connection tests). Linode is pretty stable if not very exciting, Vultr is "better than DO", but their networks are almost always in maintenance. For a little context; I maintain IRC servers and those are currently hosted in vultr (with linked nodes in 5 regions), I notice ping spikes between those nodes often and sometimes congestion which drops users. (IRC is highly stateful TCP sessions). I've only known two truly good networking suppliers, GCP (and their magic BGP<->PoP<->Dark Fibre networks) and.. Tilaa.. (which is only hosted in Netherlands.. which is why I can't use them for my global network)
        dilyevsky 6y ago
        Awesome thanks for info. For gcp i notice occasional unavailability on the order of 10s of mins every quarter or so. That’s VM networking. Their loadbalancers are a different story as they are a complete crap
        dijit 6y ago
        This is very much not my experience, do you have any more information? Any particular regions? Are you certain it's not a local ISP? (I used to run an always online video game and we had a LOOOOOT of connection issues from "Spectrum internet" on all of our servers including GCP ones.)
        dilyevsky 6y ago
        Answering here bc bottom post is locked for some reason - east1 occasionally disconnects from other regions. That is definitely within google backbone. Central-1 seems worse tho. If it’s less than an hour they dont bother with the status page. For loadbalancer its very much by design as they randomly send you rst when google rolls them for upgrade and in some other cases (I’m working on a blog post on this). Google support recommendation is to retry (foreals)
    - tgtweak 6y ago
      Internally maybe, but AWS external connectivity is so shakey I wonder if it's intentional. From a cisco ASA firewall to AWS vpg - there is weekly and even daily issues with things either plain timing out or experiencing latency spiking on AWS side. Cisco is not a small vendor or use case for creating a vpn bridge. We moved from this to cloudconnect - essentially a network connection directly to aws. Same instability issues only they moved to weekly/monthly instead of daily/weekly. This was from a provider that had fiber to AWS. I will also say that when putting particular load on AWS's internal services from a single ec2 instance you can often see requests that fail (s3 being the one I see the most). I can't speak on the merits or lack thereof on Linode, but I will say vultr and DO have - by comparision to AWS - leagues better public wan stability. I've had ssh sessions open for months on both and that includes to some POPs in europe. I'm comparing AWS which is <20ms away to something overseas... for reference that's about 20 additional hops and trans-atlantic fiber in between.
dilyevsky 6y ago
There’s also packet (now equinix metal) that gives control over l2 and have nice things such as ibgp. I think vultr may do too but their docs are poor and support was uncooperative
marvinblum 6y ago
This. I switched to Hetzner Cloud and reduced costs from ~$200/month on GCP to about ~20€/month. For comparison: 1cpu 2gb ram cost about $15 on GCP and 2,50€ on Hetzner, 1gb outgoing traffic ~0,12€ -> 20TB per machine included on Hetzner and 1€ per additional TB. I'm still using S3 and GCS to store files because it's convenient and relatively cheap.
tasogare 6y ago
This. And also AWS showed they can delete your business infrastructure unilaterally, without notice, on a whim so it’s very risky to use them.
Aeolun 6y ago
AWS has like 10 times more regions than any of the alternatives. If you want something close to home AWS is best.
- malinens 6y ago
  very very few need so many regions to choose from
  - oblio 6y ago
    Unless they're you know, not close to the main regions. That's why cloud providers have so many regions, people actually live in many parts of the world. Despite popular belief, the vast majority of the world's population does not live in the Bay Area (or even the US).

ithrow 6y ago

IME, the network of AWS is much better than that of DO, Linode.

dilyevsky 6y ago
How so?
- ithrow 6y ago
  Less hiccups and downtime. It's faster and with better latency to other third party services. Superior internal control. Ex: In, linode a private IP address gives EVERYONE on the same data center access to your Linode server. Also, last time I used them they didn't have a Firewall.
  - dilyevsky 6y ago
    Right linode is basically old school dedicated servers afaik but DO should be in different class
    - dijit 6y ago
      AWS networking isn't _great_ but it's decidedly better than DO (which is actually the worst of those listed based on my own TCP connection tests). Linode is pretty stable if not very exciting, Vultr is "better than DO", but their networks are almost always in maintenance. For a little context; I maintain IRC servers and those are currently hosted in vultr (with linked nodes in 5 regions), I notice ping spikes between those nodes often and sometimes congestion which drops users. (IRC is highly stateful TCP sessions). I've only known two truly good networking suppliers, GCP (and their magic BGP<->PoP<->Dark Fibre networks) and.. Tilaa.. (which is only hosted in Netherlands.. which is why I can't use them for my global network)
      - dilyevsky 6y ago
        Awesome thanks for info. For gcp i notice occasional unavailability on the order of 10s of mins every quarter or so. That’s VM networking. Their loadbalancers are a different story as they are a complete crap
        dijit 6y ago
        This is very much not my experience, do you have any more information? Any particular regions? Are you certain it's not a local ISP? (I used to run an always online video game and we had a LOOOOOT of connection issues from "Spectrum internet" on all of our servers including GCP ones.)
      - dilyevsky 6y ago
        Answering here bc bottom post is locked for some reason - east1 occasionally disconnects from other regions. That is definitely within google backbone. Central-1 seems worse tho. If it’s less than an hour they dont bother with the status page. For loadbalancer its very much by design as they randomly send you rst when google rolls them for upgrade and in some other cases (I’m working on a blog post on this). Google support recommendation is to retry (foreals)
  - tgtweak 6y ago
    Internally maybe, but AWS external connectivity is so shakey I wonder if it's intentional. From a cisco ASA firewall to AWS vpg - there is weekly and even daily issues with things either plain timing out or experiencing latency spiking on AWS side. Cisco is not a small vendor or use case for creating a vpn bridge. We moved from this to cloudconnect - essentially a network connection directly to aws. Same instability issues only they moved to weekly/monthly instead of daily/weekly. This was from a provider that had fiber to AWS. I will also say that when putting particular load on AWS's internal services from a single ec2 instance you can often see requests that fail (s3 being the one I see the most). I can't speak on the merits or lack thereof on Linode, but I will say vultr and DO have - by comparision to AWS - leagues better public wan stability. I've had ssh sessions open for months on both and that includes to some POPs in europe. I'm comparing AWS which is <20ms away to something overseas... for reference that's about 20 additional hops and trans-atlantic fiber in between.

dilyevsky 6y ago

There’s also packet (now equinix metal) that gives control over l2 and have nice things such as ibgp. I think vultr may do too but their docs are poor and support was uncooperative

marvinblum 6y ago

This. I switched to Hetzner Cloud and reduced costs from ~$200/month on GCP to about ~20€/month. For comparison: 1cpu 2gb ram cost about $15 on GCP and 2,50€ on Hetzner, 1gb outgoing traffic ~0,12€ -> 20TB per machine included on Hetzner and 1€ per additional TB. I'm still using S3 and GCS to store files because it's convenient and relatively cheap.

tasogare 6y ago

This. And also AWS showed they can delete your business infrastructure unilaterally, without notice, on a whim so it’s very risky to use them.

Aeolun 6y ago

AWS has like 10 times more regions than any of the alternatives. If you want something close to home AWS is best.

malinens 6y ago
very very few need so many regions to choose from
- oblio 6y ago
  Unless they're you know, not close to the main regions. That's why cloud providers have so many regions, people actually live in many parts of the world. Despite popular belief, the vast majority of the world's population does not live in the Bay Area (or even the US).