Building a Billion User Load Balancer [video] • Hacker News Board

Re: Building a Billion User Load Balancer [video]

#51

Post by skuhn » Tue, Jan 10, 2017, 6:25 PM UTC

bogomipz wrote: post #50

skuhn wrote: post #49

Earlier quoted context omitted.

High frequency trading is Solarflare's original market (and Mellanox has their InfiniBand market), but both of them are becoming more and more common in the commodity server market as well. Particularly since Intel dropped the ball on 40G (and REALLY dropped the ball on 25/100G), and Broadcom is out of the adapter market, there is a void that other vendors are filling. When you're spending $30,000 on a server, it doe…

I totally agree. It sounds like their price points @ $1,200 have come down as well. Did Intel just never get 25/100G card to market?

Not yet. They launched 40G a while ago, but some issues have kept them from the same kind of dominance they had with 10G.

25G is supposedly coming soon-ish, but 100G is still 1-2 years away. It's going to be hard to compete with vendors who shipped their first products in 2015/2016.

Meanwhile they have a 100G OmniPath adapter, but who cares?

Re: Building a Billion User Load Balancer [video]

#52

Post by bogomipz » Wed, Jan 11, 2017, 12:29 AM UTC

skuhn wrote: post #51

bogomipz wrote: post #50

Earlier quoted context omitted.

I totally agree. It sounds like their price points @ $1,200 have come down as well. Did Intel just never get 25/100G card to market?

Not yet. They launched 40G a while ago, but some issues have kept them from the same kind of dominance they had with 10G. 25G is supposedly coming soon-ish, but 100G is still 1-2 years away. It's going to be hard to compete with vendors who shipped their first products in 2015/2016. Meanwhile they have a 100G OmniPath adapter, but who cares?

>"but some issues have kept them from the same kind of dominance they had with 10G."

Oh interesting. Can you elaborate on those issue are with 40 Gig cards?

Re: Building a Billion User Load Balancer [video]

#53

Post by noponpop » Wed, Jan 11, 2017, 3:28 PM UTC

justinsaccount wrote: post #40

noponpop wrote: post #16

Interesting that Linux kernel performance (ipvs) is acceptable at l4 vs something like dpdk. I guess you just overcome the limitation by increasing the number of l4 instances load balanced by ecmp. Fun to see DSR in use. Also interesting to see that all the inherent problems with geolocation via gslb (DNS client IP is not the same as the real client IP) don't wind up being a big problem apparently. This seems to be a…

> DNS client IP is not the same as the real client IP see https://developers.google.com/speed/public-dns/faq "I've read claims that Google Public DNS can slow down certain multimedia applications or websites. Are these true? ... To help reduce the distance between DNS servers and users, Google Public DNS has deployed its servers all over the world. In particular, users in Europe should be directed to CDN content serv…

But that's google, I can tell you practically speaking many ISPs don't do this.

Client subnet extension would be very nice.