Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

cloud.google.com

Post reply on HN
Print view View original

101–110 of 158 posts

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#101

Post by mgaunard »

mintplant wrote: post #17

Oh, I was hoping this would be something built more directly over Ethernet, rather than on top of UDP/IP (if I'm understanding the layer diagram correctly). I've been working with Ethernet devices a lot lately, using the network as a communication bus, essentially. I find that there's a lot of complexity that we simply don't need: ARP, DHCP, DNS... So many points of failure. We know all the devices on our LAN and the…

All of cloud/web tech is about reinventing the wheel at upper layers. These days they're rebuilding everything on top of HTTP.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#102

Post by caskstrength »

smokefoot wrote: post #22

I don’t understand networking all that well. Is it interesting that the telcos and non-tech companies are moving away from specialized hardware toward software defined networks while the hyperscalers are using hardware acceleration?

> Is it interesting that the telcos and non-tech companies are moving away from specialized hardware toward software defined networks while the hyperscalers are using hardware acceleration?

Their SDN implementations are also hardware accelerated.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#103

Post by MichaelMoser123 »

mintplant wrote: post #17

Oh, I was hoping this would be something built more directly over Ethernet, rather than on top of UDP/IP (if I'm understanding the layer diagram correctly). I've been working with Ethernet devices a lot lately, using the network as a communication bus, essentially. I find that there's a lot of complexity that we simply don't need: ARP, DHCP, DNS... So many points of failure. We know all the devices on our LAN and the…

There are systems like this: Fibre Channel has it's own data link layer, actually they do reliability at the data link layer! I think InfiniBand is similar to this respect.

Actually it's interesting that google didn't choose any of these, for their high bandwidth storage needs. They have the money to do their own thing, but why should they?

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#105

Post by camtarn »

wayfinder wrote: post #80
allanrbo wrote: post #69

Earlier quoted context omitted.

I think op was just saying as a bus on a local network for some very specialized application.

But you can do that already. DNS, ARP, TCP/UDP, and IP are all completely optional. Just write a program to send and receive Ethernet frames. I believe it doesn’t have to contain IP.

Indeed - see industrial protocols such as EtherCAT or Powerlink, which use Ethernet frames with their own non-IP protocol on top.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#106

Post by eecc »

mgaunard wrote: post #101
mintplant wrote: post #17

Oh, I was hoping this would be something built more directly over Ethernet, rather than on top of UDP/IP (if I'm understanding the layer diagram correctly). I've been working with Ethernet devices a lot lately, using the network as a communication bus, essentially. I find that there's a lot of complexity that we simply don't need: ARP, DHCP, DNS... So many points of failure. We know all the devices on our LAN and the…

All of cloud/web tech is about reinventing the wheel at upper layers. These days they're rebuilding everything on top of HTTP.

Heh, what a coincidence. Just about yesterday I was chatting with a friend about the similarities between Service Mesh and the ESB of yore.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#107

Post by jacquesm »

stonogo wrote: post #95
Double_a_92 wrote: post #90

Is it just me getting older / less smart, or did articles about products really start to sound like a jumbled mess or buzzwords lately? What is "Hardware transport", what is "the ecosystem"? And then there is dozens of random products and technologies that I've never heard of... This sounds more like a humble brag, than an article trying to inform people about technologies that might actually be useful to them.

"Hardware transport" is kind of a misnomer, because this is a networking protocol. It just happens to be a networking protocol that requires hardware acceleration on the NIC. We already had that in things like Infiniband and Omnipath, but Nvidia bought one and Intel rugpulled the other. Meanwhile ethernet has been approaching parity in terms of throughput, but TCP introduces unpleasant latency, so this is Google's NI…

> It's something of a rite of passage for a company to convince itself that building this sort of thing in-house is necessary, and that it will revolutionize high-performance computing in all the ways that previous, nearly-identical projects have not.

The same happens with:

- databases

- encryption

- operating systems

- frameworks

- programming languages

I've seen this so many times by now it stopped being funny.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#108

Post by gertrunde »

genmud wrote: post #46
mintplant wrote: post #17

Oh, I was hoping this would be something built more directly over Ethernet, rather than on top of UDP/IP (if I'm understanding the layer diagram correctly). I've been working with Ethernet devices a lot lately, using the network as a communication bus, essentially. I find that there's a lot of complexity that we simply don't need: ARP, DHCP, DNS... So many points of failure. We know all the devices on our LAN and the…

But you can't route MACs... you can't associate domain names to MACs... Maybe for a LAN that's fine, but for connecting devices between broadcast domains, it just doesn't cut it.

You could view the vxlan/evpn/mp-bgp combo as routing MACs... sort of :)

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#109

Post by jeffbee »

jacquesm wrote: post #107
stonogo wrote: post #95

Earlier quoted context omitted.

"Hardware transport" is kind of a misnomer, because this is a networking protocol. It just happens to be a networking protocol that requires hardware acceleration on the NIC. We already had that in things like Infiniband and Omnipath, but Nvidia bought one and Intel rugpulled the other. Meanwhile ethernet has been approaching parity in terms of throughput, but TCP introduces unpleasant latency, so this is Google's NI…

> It's something of a rite of passage for a company to convince itself that building this sort of thing in-house is necessary, and that it will revolutionize high-performance computing in all the ways that previous, nearly-identical projects have not. The same happens with: - databases - encryption - operating systems - frameworks - programming languages I've seen this so many times by now it stopped being funny.

Are you suggesting that it would have been better for Google to use off-the-shelf databases etc? Because at their scale it seems clearly necessary to bring that in-house.

Re: Google opens Falcon, a reliable low-latency hardware transport, to the ecosystem

#110

Post by jacquesm »

jeffbee wrote: post #109
jacquesm wrote: post #107

Earlier quoted context omitted.

> It's something of a rite of passage for a company to convince itself that building this sort of thing in-house is necessary, and that it will revolutionize high-performance computing in all the ways that previous, nearly-identical projects have not. The same happens with: - databases - encryption - operating systems - frameworks - programming languages I've seen this so many times by now it stopped being funny.

Are you suggesting that it would have been better for Google to use off-the-shelf databases etc? Because at their scale it seems clearly necessary to bring that in-house.

At Google's scale a lot of the ordinary limitations do not apply. Unfortunately many companies believe that because Google does it it must be good. It's cargo cult reasoning and the result is endless NIH projects.

As for Google's 'Falcon' project: it smacks of NIH to me, but maybe their use cases are specific enough that none of the off-the-shelf bits were usable.

Post reply on HN