Drunkard's Walk Forums

Full Version: Size of Interweave nodes
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2

HRogge

I think we should discuss about the size and abilities of Interweave nodes.

I have been working on a tech gadget for FTL communication. Up to this point I used the wiki information that Interweave nodes are bulky and large, but people on the IRC told me that this information was outdated and they were as large as the plot demanded. After checking with them, I changed the Interwave wiki entry to reflect this, which was reverted by Robkelk.

So I am a little bit confused what is the current consensus about Interweave nodes.
I think part of it is that "plot" demands, mostly, that the "Interwave" nodes be similarly as hard to move around as the Internet Backbone on earth.

The actual hardware that does the "oomph" can be as small as an iPod slathered in Mnemosyne's Honey, but the connections to the rest of the Interwave and maintenance supplies, etc. take up a lot of space.
''We don't just borrow words; on occasion, English has pursued other languages down alleyways to beat
them unconscious and rifle their pockets for new vocabulary.''

-- James Nicoll
What is an interwave node?

From what I gathered, the main nodes where huge for the sole reason that they had to broadcast and receive terabytes of information a second, from many thousands of individual connections. They're the connection hubs on which the whole system is based. The individual user at the end of the station doesn't need to be able to handle more than their own needs. Compare the little 3G dongle you can get from most mobile networks, with some of the larger mobile phone towers... (For which the actual base transceiver is also surprisingly small, most of it's just the broadcasting equipment.)

A cyborg could have the little USB dongle thing inside them easy enough, but only really be able to hook into nearby nodes at short ranges. (Jet can't reach the geostationary equipment from Earth's surface, she has to use a large wave enabled antennae dish at a Motel).
________________________________
--m(^0^)m-- Wot, no sig?
That's a pretty good analogy, Dartz.

Real-world network nodes are, at minimum, an http://en.wikipedia.org/wiki/Rack_unit]8U box - that's 19"×19"×14", plus an external power supply. They need to be this large because of all the connectors that plug into them. And that's for internal networks - even the smallest Internet nodes are made up of multiple 8U boxes.

Certainly, someone in Fenspace can handwave a terminal to something the size of a lapel pin (and no doubt the Trekkies do that on a mass-production basis), but that isn't a node. It probably isn't even Interwave (although one should never rule out the Mads).
--
Rob Kelk
"Governments have no right to question the loyalty of those who oppose
them. Adversaries remain citizens of the same state, common subjects of
the same sovereign, servants of the same law."

- Michael Ignatieff, addressing Stanford University in 2012

HRogge

I think nobody is assuming that you can put a full blown backbone node which handles multiple terabit/s streams to other nodes lightminutes away into a small package.

The interesting question is what is the minimum size for a FTL node that connects a single local network with typical DSL speed (several megabits/s) to a close (1 lightsecond ? 1 lightminute ? even more ? less ?) full blown backbone node ?

The wiki seems to suggest that this minimum size is still "huge"... too large for converted cars, even too large for some spaceships.
My impression of the Interwave size issue was that the antenna was both large and complicated, hence the need for a LOT of space for them and there support equipment. More space than anything less than a large SV could afford. Around the Delta Pavonis mission the Mk II came around (the main impetus was A.C. wanting more bandwidth for KoFen sessions, but work had been proceeding for a while) with the basic system (not a node, but a reasonable FTL com system) shrunk to a quarter to something that could fit in a mid-sized room, with most of that being the antenna assembly (my imaginings had an antenna about 2m*2m*2.5m plus about 2 15u transceiver coils and a 2u controller with fibre connections).

The main problem with needing the size is the range. To get equivalent ranges on various FTL com systems you need a system larger than a comparable radio system unless you're more than 2 lightseconds away. Interwave got its penetration because it could be scaled to handle the traffic that tended to be generated. Plus enhanced radio is much cheaper.

My opinion only.
What I expect is that it's a combination of size if you want any real range, and basically positioning and orientation. In some cases, yes, you can easily move the node... but then you have to spend days or weeks, depending on the node and where you moved it to, getting the thing to talk to the other nodes again. And the hardware and software necessary to keep it connected real time while being moved would likely add a factor of size and complexity as well, certainly more than most Fen would be willing to fuss with for a node that's normally relatively stationary.

Many ships (such as the Blue Midget fleet of JMC) may ship with Interwave nodes onboard... but those nodes are also not designed explicitly to be 'backbone' nodes, if only because they're in more or less a constant state of motion. They can require at least one computer-steerable transmitter and receiver, or a decent fixed antenna on small installations such as cars. And the connection is not going to be reliable enough to get the large amount of bandwidth required to be a backbone system.

JMC base facilities, such as Green Planetoidy and the 'main' office in Serenity Valley, do have true backbone nodes, but they are usually regarded as being immobile because of their relative complexity, being tied into multiple pre-aimed emitters and receivers on the outside surfaces of the station, having multiple redundancy, and whole set of 19U hardware racks loaded to the gills with 'waved hardware, which can also be hard to justify in terms of space in a moving spacecraft.
--

"You know how parents tell you everything's going to fine, but you know they're lying to make you feel better? Everything's going to be fine." - The Doctor
Quick thoughts

A Node would be big.You've got thousands upon thousands of connections going through it. You'd need a big array of transmitters feeding down into some really power hungry computer hardware routing it out. These things have to be big, because they have to serve a wide area of space. You'd also have a large amount of computing power going solely towards maintaining some reasonable lookup system, so you can tell where to route individual data to. These also have to have the largest range, potentially to the nearest stars (Though this can be somewhat directional). Each node has it's own broadcast frequency.

Beneath that, you have your 'towers'. These could be reliably mounted aboard ships or satellites. A ship-based system only needs to handle the data from within that ship and within a certain range around the ship. It only needs range enough to be able to reach the nearest node, (or another tower) which will usually be within 5-15 light minutes. The towers will also be able to point directly at the node, since the main node positions are on reasonably fixed structures, so won't need as much power to broadcast back to the node, as the node does to broadcast to them. Each 'tower' will transmit on a different frequency, such that no towers connected to the one node are broadcasting on the same frequency. Some of these may move about between nodes occasionally, and may be required to shift their broadcast frequency. As an example SS Ciara's system has a receive/broadcast range back to a main node of about 10 light minutes, and would be considered average. It can work from the ground, and is normally used to provide interwave coverage for her crew, and for her two boats and anyone else within a few light seconds.

At the bottom end, you've your interwave receiver dongle, which is solely capable of connecting to 'tower'/nodes withing a 20-30k KM. It handshakes with the nearest 'tower' which tells the entire network where it is. It can handle it's own data, and not much elses. And Example would be Jet Jaguar's onboard system.... which doesn't even connect to geostationary satellites from Earth orbit. They transmit and receive on the frequency of the 'tower' they connect to, and can change this frequency as they move about.

Edit: Bah, double Ninja'd.
________________________________
--m(^0^)m-- Wot, no sig?
Quote:Edit: Bah, double Ninja'd.
Don't worry about that - saying something in a different way often provides new insight.

It looks like we've figured out the difference between a terminal, a repeater, and a node now. (Yay.) We're at the point where we're discussing the sizes of each of those...
--
Rob Kelk
"Governments have no right to question the loyalty of those who oppose
them. Adversaries remain citizens of the same state, common subjects of
the same sovereign, servants of the same law."

- Michael Ignatieff, addressing Stanford University in 2012

HRogge

And we need a "typical" range between a terminal and the repeater. Because of lightspeed delay the repeater makes only sense if it has thousands of km range... if not the difference between Interwave and Radio will be too low.
Not to mention how the Wizards' biomodded-owl post system manages to interface with it -- without a noticeable drop in transmission speed.
-- Bob
---------
Then the horns kicked in...
...and my shoes began to squeak.
Thats just the miracle of handwavium in action Bob. what no one in universe realises is that the owls actually are able to teleport, but only when no one can see them do it.
Hear that thunder rolling till it seems to split the sky?
That's every ship in Grayson's Navy taking up the cry-

NO QUARTER!!!
-- "No Quarter", by Echo's Children
Quote:Bloody inconvenient, is what it is, Postal Owl H3D-W16 grumbled to herself as she closed her eyes to initiate her teleport.

Like that?
''We don't just borrow words; on occasion, English has pursued other languages down alleyways to beat
them unconscious and rifle their pockets for new vocabulary.''

-- James Nicoll
RFC1149, replace the pigeons with owls?
________________________________
--m(^0^)m-- Wot, no sig?
All I can add is there's a reason the Wizards' biggest Interwave node is on asteroid Hedwig...
--
Rob Kelk
"Governments have no right to question the loyalty of those who oppose
them. Adversaries remain citizens of the same state, common subjects of
the same sovereign, servants of the same law."

- Michael Ignatieff, addressing Stanford University in 2012
Very large ships like the Tyger Tyger may even has a system setup for relaying interweave traffic from smaller craft around them. Sort of like a mobile hotspot.
_______________________________________________________________
Characters
Sabre Fang
Dakota

HRogge

So the consensus about Interwave nodes go along the line Dartz suggested above ?

What do you think about this as a draft ?
  • Interwave Backbone Nodes
    Large or even huge installations that can (depending on size) handle one or multiple point to point connections with other Backbone Nodes. They have a range measured in lightyears and transport terabits of data per second. Only the larger ships can carry a small Backbone Node
  • Interwave Cell Towers
    These towers are waved 3G/4G phone equipment that allows specially waved client devices within a large area (3000km to 30000km, depending on size of the tower?) to connect to the Interwave without lightspeed lag. Interwave Cell Towers only work outside atmosphere (on planets, the line of sight is too short to make FTL to a cell network very useful).
  • Clients Devices
    Clients log into the nearest Cell tower to allow access to the Interwave through the tower. Depending on the size an capabilities of the client device, they allow a bandwidth of several to hundreds of megabits per second.

Many Interwave Clients also have point to point sublight radios built in.
There are some smaller jamming systems to block the Client devices or the local Interwave Cell Tower, jamming a Interwave Backbone Node needs much more power (and larger jammers).
Looks good to me...
--
Rob Kelk
"Governments have no right to question the loyalty of those who oppose
them. Adversaries remain citizens of the same state, common subjects of
the same sovereign, servants of the same law."

- Michael Ignatieff, addressing Stanford University in 2012
HRogge Wrote:So the consensus about Interwave nodes go along the line Dartz suggested above ?

What do you think about this as a draft ?
  • Interwave Backbone Nodes

    Large or even huge installations that can (depending on size) handle one or multiple point to point connections with other Backbone Nodes. They have a range measured in lightyears and transport terabits of data per second. Only the larger ships can carry a small Backbone Node
  • Interwave Cell Towers

    These towers are waved 3G/4G phone equipment that allows specially waved client devices within a large area (3000km to 30000km, depending on size of the tower?) to connect to the Interwave without lightspeed lag. Interwave Cell Towers only work outside atmosphere (on planets, the line of sight is too short to make FTL to a cell network very useful).
  • Clients Devices

    Clients log into the nearest Cell tower to allow access to the Interwave through the tower. Depending on the size an capabilities of the client device, they allow a bandwidth of several to hundreds of megabits per second.

Many Interwave Clients also have point to point sublight radios built in.

There are some smaller jamming systems to block the Client devices or the local Interwave Cell Tower, jamming a Interwave Backbone Node needs much more power (and larger jammers).
With a "Hotspot" being a form of mobile Cell tower attached to a very large ship with limited connections. Also connection range for client devices might be from 10km to 100km as most of the broadcast power is going to connections to other towers.
_______________________________________________________________
Characters
Sabre Fang
Dakota

HRogge

Dakota Wrote:With a "Hotspot" being a form of mobile Cell tower attached to a very large ship with limited connections.
Yes...

Quote:Also connection range for client devices might be from 10km to 100km as most of the broadcast power is going to connections to other towers.
why wave something to be FTL at a range of 10-100 km ? the signal delay would be 1-3 microseconds for normal radio.
HRogge Wrote:
Dakota Wrote:With a "Hotspot" being a form of mobile Cell tower attached to a very large ship with limited connections.

Yes...

Quote:Also connection range for client devices might be from 10km to 100km as most of the broadcast power is going to connections to other towers.

why wave something to be FTL at a range of 10-100 km ? the signal delay would be 1-3 microseconds for normal radio.
It can be setup to take radio as well, but wouldn't it just be easier to accept close range client connections via interweave and boost the signal out to other towers?
_______________________________________________________________
Characters
Sabre Fang
Dakota
I think the question is more like "What would the burden (power consumption, technological complexity, vulnerability to failure) be to have proper FTL be restricted to Node-Node, Node-Tower, and Tower-Tower connections, and have the Terminal-Tower and Terminal-Terminal connections be strictly sublight, despite using the Interwave network stack?" Remember, Interwave is a Fennish extension to existing Internet protocols. The Internet Protocol Suite has four layers: Application, Transport, Internet, and Link. Hardware is entirely within the Link Layer. Assuming the InterWave Protocol Suite follows the same pattern, there is absolutely nothing that prohibits entirely sublight Interwave networks.

HRogge

I think all of Fenspace use the Interwave protocols everywhere... the question is how far does the faster than light communication go towards smaller users.
Reduced latency for important things, such as online gaming..... of course.
________________________________
--m(^0^)m-- Wot, no sig?
Which one isn't likely to do from a handheld computer (or a wristcomp)...

I can see arcades / gaming parlors being popular in the Main Belt, because one-person ships only have regular comms and nobody wants their characters to be fragged because of lightspeed delays.
--
Rob Kelk
"Governments have no right to question the loyalty of those who oppose
them. Adversaries remain citizens of the same state, common subjects of
the same sovereign, servants of the same law."

- Michael Ignatieff, addressing Stanford University in 2012
Pages: 1 2