LiFi seems a poor fit for most home use. Maybe a point to point link to the detached garage, or jumping a wired link across a hard to route room? Occlusion seems a non-trivial problem for most wireless devices.
A reliable VR or home theater setup with ultra low latency might convince me it's worth the effort, but that would still involve some wiring.
Almost had a heart attack: I thought you are really asking nerds about life. Until I googled it...
Nobody said all devices in the network need 10Gbit. Most devices sure don't. Just make a "VIP" 10GBit network for a few devices and leave the "pleb" devices on 1GBit.
That's obvious, but economically the VIP area could only contain 4 seats (a switch only makes sense for >=3 devices to begin with, and at least I wouldn't feel comfortable filling up all of them on day one which would entail having to buy another switch should VIP device 5 appear some time later) - which is very limiting. More reasonable 8 ports quickly blow the (switch-) budget.
EDIT: I don't want to derail this thread even further, but yes, four 10 GBit/s ports wouldn't be anything I'd pay significant money for, why?
Assume I had only two 10 GBit/s systems
--> no need for a switch, a peer-to-peer connection (in addition to plain boring 1000BASE-T to connect to the rest of the world) would be sufficient.
Assume I had three 10 GBit/s systems
--> here we are slowly entering switch territory, functionally the best choice - but still at cross-roads (peer-to-peer might still come out cheaper (think used infiniband or SFP ethernet cards with DAC cabling)
One free switch port, not much extensibility (a 10 GBit/s card to retrofit isn't that exotic, and in a quantity of one semi-affordable, depending on how much use the system really gets and how much it would benefit from the speedup).
Assume I had four 10 GBit/s systems
--> all 10 GBit/s ports in use, no free ports left, switch too small.
The question for me is, what would I invest real cash in (it's different if I'd basically get the 10 GBit/s ports 'for free', as part of a device (switch) I'd have to buy (for other reasons) anyways. As an 'enthusiast', I'm certainly willing to 'lie' to myself to some extent (overexaggerate my needs and paint it in rainbow colours) - but only in only in a limited and specific way…
Likewise I have no motivation to 'invest' in 2.5GBASE-T or 5GBASE-T, fine if it comes 'for free' with future mainboards (still wouldn't pay real cash for a switch that couldn't do native 10GBASE-T/ SFP at the same time) - but I'm not replacing my 1000BASE-T infrastructure for that.
For me personally, the absolutely minimum port count in a switch is eight (and to be perfectly honest, I'm much more comfortable with 16-28 ports, at least for 1000BASE-T, but hey - prices dictate what's reasonable and what's not, for 10GBASE-T in particular) - and that's already 'uncomfortable' (in other words, this forces me to plan ahead - not lay cables/ connect as convenient).
To clarify this further, I currently have two systems which would really benefit from >>1 GBit/s LAN speeds, and a third that would profit from faster connections as well (all systems regularly seeing LAN file transfers in excess of 150-700 GB) - then there is wifi6/ wifi6e, it's not that far out that my router/ AP wants >=2.5GBASE-T connectivity as well (not for its WAN, but for its WLAN; intel AX210 wireless cards are already readily available and affordable).
Do you really need more than 4 VIP seats anyway? I'm assuming relatively intelligently designed switches that have at least one gbit port so you don't waste a 10gbit port for uplink to the router/internet and the rest of the low-speed LAN. (also the other guy has bought one of those)
With 4 seats you can connect a home server/NAS and three PCs that do things that can actually benefit from 10Gbit.
For a "home budget", I'm not sure 10G ports would make the cut for things I'd recommend paying for.
For an "enthusiast", I'd look for 1-4 ports. Something that can connect a multi-user device (like a home server) or can provide enough connectivity to hang a separate 10G network off the primary home network for future expansion.
Past that I believe it starts getting into "home lab" territory for hobbyists or "home business" for professional needs. If a home is regularly using 10GBit traffic between 4+ devices, I'd say they are well outside the typical home use case and the justification for the cost becomes either "because it brings joy" or "because it pays for itself".
For CeroWrt II, I think it's worth targeting hardware that has at least a 10G port.
I would like to debloat india. There, a cup of coffee is 25 cents. Other profit margins and costs are similar in scale, so I was leaning towards a cheap platform - gl300-like at best, with a little more flash and memory, as being suitable for that market.
nods I can get behind that. Do you plan on trying to target power consumption? Packets per watt perhaps? ^-^
heh. Ideally I'd like to get away from mips. It's a dead end now, as much as I love the ath9k. I really haven't kept track of the market (?) and hope that mt76 stuff was dropping down in price. The gl300 was a pretty good box...
To the 10Gbit point, showing how cake can improve a 10Gbit link - on some cheap platform - is also on my mind. I mostly fear we have to make that big move to xdp and a xpf firewall compiler to make progress at these speeds.
The mikrotik folk have finally added fq_codel and cake across their product line in the beta, but there's no means to dynamically reconfigure it. They do make some interesting looking hw that can be reflashed to openwrt.
I would really love a proof of concept that used libreqos on openwrt, on a wndr3800.
While I am doing the blue sky thing, NDN still lives: https://named-data.net/publications/ndn-hackathons/
We have fq_codel support in our community nss builds for the IPQ806x, which is accelerated by the nss cores.There is pretty much no cpu load when shaping 1Gbps. I added a very basic script for that, in order to be selectable in webif:
It could use some work, but I don't have much time lately.
It's also worth asking if there is something simpler and lower computation than cake which could help these speeds. Like hashing to bfifos below a qfq, and a TBF on top maybe. Even 1000 MTU packets takes 1.2ms at 10G speed so you could have a lot of slop compared to optimal without much bloat.
@kong it's proof that it's actually working properly that I'm after. Dropping packets.
Well I used flent to check if it is working, it is not as good as cake but it is ok.
@kong thx for testing with flent. Got the flent.gz files anywhere? Flent mostly shows fq working not codel working unless you stare at it hard.
Also the burst parameter seems to be hard coded very, very large.
A good test is to knock it to 200Mbit and use
--step-size=.05 --socket-stats --te=upload_streams=4 tcp_nup
For science! I have a nearly religious love of drop head queuing vs tail drop, but showing any benefits and 10gbit performance on cheap hardware of anything but a FIFO would be good.
Unfortunately I deleted them, but I can run the tests again tomorrow. But I only have a 100Mbps ISP connection, for higher speeds I would have to setup some local server on the wan side of the router.
Not sure if that comes close to a real world scenario.
well, my concern is that there's problems above 100Mbit. Does the nsstbf work at 900mbit? inbound?? It would be cool if it worked in bound.... It's a very real world scenario to have a server "right there" on the wan side....
But: Certainly verifying that it worked at 100Mbit would be a start! But the regular sqm-scripts use a scalable burst and htb quantum you don't have....