4 ms·

I really do wonder why Azure is so slow. You have to optimize somewhat aggressively to squeeze average performance out of Azure SQL, for instance.

vorticalbox 7y ago
Maybe because Windows updates and xbox love run through the network?
- weberc2 7y ago
  Pretty sure all of Netflix runs on AWS and I've got to think that's several orders of magnitude more traffic than Windows updates or xbl.
  - jlgaddis 7y ago
    FWIW, the video streaming part of Netflix does not run in AWS (but, AIUI, pretty much everything else does).
    - weberc2 7y ago
      Oh, TIL I guess. Thanks for the correction.
  - ofrzeta 7y ago
    Netflix hosts video content at ISPs' sites: https://openconnect.netflix.com/en/ https://openconnect.netflix.com/en/
    - slenk 7y ago
      Edge content yes, but everything else is in AWS
      - GhettoMaestro 7y ago
        The edge traffic makes up 99%+ of the traffic in question.
        slenk 7y ago
        Do they publish how many ISPs participate? I am sure the big ones do but who knows if local Fiber is connected to someone who is behind an OpenConnect
        sanxiyn 7y ago
        Non-participating ISPs get it from participating ISPs. Basically no one hits AWS.
        ivnubinas 7y ago
        Dominican here studying Telecommunications. I was giving a tour to the Datacenter of the 3rd largest telecom company in my country. They were using the Netflix box. I was giving a tour to the 4th largest telecom company, they too had the Netflix box. I was told the 1st and 2nd largest telecoms in the country also had the box.
        GhettoMaestro 7y ago
        My understanding is that Netflix will pretty much subsidize the cost of the CDN box to nearly any ISP above a certain minimum traffic level ingress from Netflix. Great model to attack the incumbent [transport / transit] carrier costs - end run around the middle-men and drop your CDN boxes directly at the edge (last-mile) ISPs.
        walrus01 7y ago
        ISP here, the netflix cache appliance boxes (a single 1RU server is the minimum) are free, but you have to provide rack space, power and cooling for them. For an ISP in a place like a caribbean island it's a great idea because their upstream/IP transit rates are significantly higher than in the US48 states.
jermaustin1 7y ago
Back in 2012 or 2013 I built an auto-scaling WordPress hosting off of Azure. It worked OK, better than shared hosting, but the $15 base VM at Azure (it was a cheaper time back then) couldn't compete with the base $10 droplet at Digital Ocean (it was a more expensive time back then). On top of that Digital Ocean's API was so much better than Azure. So I spent a week and switched to Digital Ocean. Needed to scale the instances less, cost less, so more profit! Unfortunately after 2 months of development, and half a year of marketing it, we only managed to attract a handful of customers, and had to shut it down.
- twobat 7y ago
  Why did try to scale when you had no customers?
  - dx034 7y ago
    If you want to offer a completely automated process, auto-scaling is important. Otherwise it'll fail as soon as the first customers come in.
  - jermaustin1 7y ago
    I guess I wasn't perfectly clear. The technology was to scale the customer's wordpress installation when certain metrics were met CPU or RAM > 75%, clone droplet and load balance. The technology was basically an a beefy server for NFS, a beefy MySql server, and a beefy load balancer. When you started hosting with us, you paid $50+/mo and got what felt like a standard WordPress install on a dedicated digital ocean droplet, when metrics showed that you were hitting the scale threshold, we would create a new droplet run an initialization script to mount NFS, set up the nginx server, and then attach it to the loadbalancer. When the combined CPU or RAM droped below the initial scale threshold, it would drop one VM from the load balancer at a time keeping it provisioned for an hour in case it needed to scale again. Worked amazingly well. Sure the load balancer was a single point of failure, but in the 10 or so months that it was running, we never had a second of outage based on pinging the websites.
    - 7y ago
thu2111 7y ago
It's massively, massively overcommitted. They managed to actually run out of hardware in a whole US region recently, even though of course most customers had lots of unused quota. A typical Azure VM must be extraordinarily busy given some of the latencies and speeds I've seen on them.

vorticalbox 7y ago

Maybe because Windows updates and xbox love run through the network?

weberc2 7y ago
Pretty sure all of Netflix runs on AWS and I've got to think that's several orders of magnitude more traffic than Windows updates or xbl.
- jlgaddis 7y ago
  FWIW, the video streaming part of Netflix does not run in AWS (but, AIUI, pretty much everything else does).
  - weberc2 7y ago
    Oh, TIL I guess. Thanks for the correction.
- ofrzeta 7y ago
  Netflix hosts video content at ISPs' sites: https://openconnect.netflix.com/en/ https://openconnect.netflix.com/en/
  - slenk 7y ago
    Edge content yes, but everything else is in AWS
    - GhettoMaestro 7y ago
      The edge traffic makes up 99%+ of the traffic in question.
      - slenk 7y ago
        Do they publish how many ISPs participate? I am sure the big ones do but who knows if local Fiber is connected to someone who is behind an OpenConnect
        sanxiyn 7y ago
        Non-participating ISPs get it from participating ISPs. Basically no one hits AWS.
        ivnubinas 7y ago
        Dominican here studying Telecommunications. I was giving a tour to the Datacenter of the 3rd largest telecom company in my country. They were using the Netflix box. I was giving a tour to the 4th largest telecom company, they too had the Netflix box. I was told the 1st and 2nd largest telecoms in the country also had the box.
        GhettoMaestro 7y ago
        My understanding is that Netflix will pretty much subsidize the cost of the CDN box to nearly any ISP above a certain minimum traffic level ingress from Netflix. Great model to attack the incumbent [transport / transit] carrier costs - end run around the middle-men and drop your CDN boxes directly at the edge (last-mile) ISPs.
        walrus01 7y ago
        ISP here, the netflix cache appliance boxes (a single 1RU server is the minimum) are free, but you have to provide rack space, power and cooling for them. For an ISP in a place like a caribbean island it's a great idea because their upstream/IP transit rates are significantly higher than in the US48 states.

jermaustin1 7y ago

Back in 2012 or 2013 I built an auto-scaling WordPress hosting off of Azure. It worked OK, better than shared hosting, but the $15 base VM at Azure (it was a cheaper time back then) couldn't compete with the base $10 droplet at Digital Ocean (it was a more expensive time back then). On top of that Digital Ocean's API was so much better than Azure. So I spent a week and switched to Digital Ocean. Needed to scale the instances less, cost less, so more profit! Unfortunately after 2 months of development, and half a year of marketing it, we only managed to attract a handful of customers, and had to shut it down.

twobat 7y ago
Why did try to scale when you had no customers?
- dx034 7y ago
  If you want to offer a completely automated process, auto-scaling is important. Otherwise it'll fail as soon as the first customers come in.
- jermaustin1 7y ago
  I guess I wasn't perfectly clear. The technology was to scale the customer's wordpress installation when certain metrics were met CPU or RAM > 75%, clone droplet and load balance. The technology was basically an a beefy server for NFS, a beefy MySql server, and a beefy load balancer. When you started hosting with us, you paid $50+/mo and got what felt like a standard WordPress install on a dedicated digital ocean droplet, when metrics showed that you were hitting the scale threshold, we would create a new droplet run an initialization script to mount NFS, set up the nginx server, and then attach it to the loadbalancer. When the combined CPU or RAM droped below the initial scale threshold, it would drop one VM from the load balancer at a time keeping it provisioned for an hour in case it needed to scale again. Worked amazingly well. Sure the load balancer was a single point of failure, but in the 10 or so months that it was running, we never had a second of outage based on pinging the websites.
  - 7y ago

thu2111 7y ago

It's massively, massively overcommitted. They managed to actually run out of hardware in a whole US region recently, even though of course most customers had lots of unused quota. A typical Azure VM must be extraordinarily busy given some of the latencies and speeds I've seen on them.