- What is Netsukuku? What's behind the scene?
- What is the performance of the network? Why should it work?
- What about the security of my packets? Everyone will be able to sniff them!
- How does a new node begin to locate any of the other nodes in the network?
- Will you provide "Internet to Netsukuku" tunnels?
- What can be done on the Netsukuku network?
- Will glibc be able to resolve names for the ANDNA system?
The Netsukuku project is based on the very simple idea of exploiting the great potentiality of the wifi connectivity, making the PCs of wireless communities act as routers and handle together an ad-hoc network even bigger than the Internet.
Netsukuku is an ad-hoc network system designed to handle massive numbers of nodes with minimal consumption of CPU and memory resources. It can be used to build a world-wide distributed, fault-tolerant, anonymous, and censorship-immune network, fully independent from the Internet. Netsukuku does not rely upon any form of backbone router, internet service provider network, or any centralized system, although it may take advantage of existing systems of this nature to augment unity and connectivity of the existing Netsukuku network.
A new Netsukuku user needs do little more than install an antenna within range of other local nodes, linking themselves into the network, and run the Netsukuku software on their computer to take advantage of it. The number of interconnected nodes can grow endlessly. If a node is out of the range of any wifi signals, a "virtual tunnel" over their Internet connection will supply the missing radio link.
Over the Netsukuku network various primary distributed services are guaranteed. One of these is ANDNA (A Netsukuku Domain Name Architecture), which is the non hierarchical and decentralized system of hostname management used in the Netsukuku network. It fully replaces the DNS and any node can register up to 256 hostnames.
For a more detailed plain-english review of the ideas behind the Netsukuku project, with a focus upon core concepts and capabilities see: netsukuku.pdf
For a technical description of Netsukuku see What is Netsukuku? What's behind the scene?
For anything else, read the main documentation
The project is well and alive.
There's no operative software available yet, but we are almost done with the first minimal release of PyNtk. It is a working Python implementation of Netsukuku and has just the basic features to create a simple Netsukuku network, along with hostname management.
We'are also strengthening the theory, researching and writing proofs for the validity and performance of the Netsukuku algorithms.
Check out our roadmap
Netsukuku sounds like "network" in Japanese, and we like Japanese stuff. Moreover, when the project started, no results could be found for "Netsukuku" on Google. Hence, choosing this name was a natural choice for us.
Netsukuku is not a p2p net built upon the Internet, i.e. overlay network. It is a physical network and it is a dynamic routing system designed to handle up to 2^32 nodes without any servers or central systems, in this way, it is possible to build a physical network separated from the Internet. Read What is Netsukuku?
Netsukuku should reasonably work. The most intuitive argument is this: Netsukuku is just trying to automate what the Internet does everyday, that is keeping routes, assigning IP addresses and resolving hostnames.
By the way, we are currently researching and writing proofs for the validity and performance of the Netsukuku algorithms.
As for now, in the documentation every claim is supported by a complete reasoning. You can find detailed explanation of the protocols, the way of working of the algorithms and how and why they have been chosen. And if the doc isn't enough, you can always communicate with the developers (mailinglist).
As a side note, there isn't any written rules forbidding contributor to develop new algorithms. If you want to help in the development, read the docs, code and contact us ;)
The first operative release of the Netsukuku software hasn't been released yet. Therefore, there shouldn't be any. Check out the website for updates.
Take out your wifi antenna, and start the Netsukuku software, then wait and tell all your neighbors to do the same.
It is up to you. Some examples of things you can do:
- help the developer writing the Netsukuku software.
- study the Netsukuku theory and algorithms. Improve it.
- expand and translate the documentation.
Check out the TODO list .
Contact us for any question or contribution.
You can call Netsukuku a "scalable ad-hoc network architecture for cheap self-configuring Internets". Scalable ad-hoc network architectures give the possibility to build and sustain a network as large as the Internet without any manual intervention. Netsukuku adopts a modified distance vector routing mechanism that is well integrated in different layers of its hierarchical network topology. The Netsukuku software is distributed and requires very few memory and computational resources: the high degree of self-similarity of its topology grants the ability to store the entire network routing table in just few kilobytes; the routing discovery algorithm consists only in the non-invasive generation and propagation of Tracer Packets. Thanks to these low requirements, it is possible to run the software inside low-performance computers, access points and embedded devices. Some relevant features that exploit the architecture offered by Netsukuku are: a distributed, non hierarchical and decentralised system of hostname management; the easy integration of P2P overlay services; an Internet tunnelling system that connects nodes which aren't physically linked; a system which enables full anonymity, hiding the source and destination of packets and encrypting them. Using wireless technologies, the users of Netsukuku can build and sustain a wide area network by sharing a small part of their PC resources, without requiring any third party specific support.
For more information see the main documentation
We're currently researching and writing proofs for the validity and performance of the Netsukuku topology and routing system. Hence, you are invited to wait the Netsukuku developers or to try find proofs yourself.
If you want to evaluate the Netsukuku protocol performance, you should take into account the following factors (in order of priority):
- Topology (how the logic structure of the network is organised)
- Dynamics of the topology: what happens when a node changes its status (see "7 Hooking phase")
- IP distribution (see "7.1 Uniform gnodes"): how the IP addresses are allocated, trying to avoid waste of space.
- Routing in high levels (see "6 Flat levels"): how the QSPN is used in the high level of the hierarchy.
- "7.2 Internal connection": what happens when the network is split in two separated nets.
- ETP: valuate its performance, etc...
The real difficulty doesn't concern the particular routing algorithm used, but lies in the network structure and organisation. For example, instead of the QSPN you can try to use a normal routing discovery method: it will always work inside a group of nodes (256).
Then you have to take into account the higher levels of the hierarchy, i.e. group of group of nodes, but here you can still try to use your favorite Distance Vector routing scheme.
For further information, you can find the appropriate documentation here.
The situation in Netsukuku isn't much different from the current Internet:
in the Internet, only the ISPs, the routers and the backbones can sniff, alter and destroy your traffic.
in Netsukuku, only the nodes belonging to the temporary route, which you are using for your connection, will be able to sniff, alter and destroy your traffic. However consider that:
- with the use of multipath your traffic is split among different routes, thus each route will be able to read only a partial portion of your traffic.
- for each connection you have a potential different route
Both in the Internet and in Netsukuku, you can always use secure protocols, such as https, SSH, SSL, and so on to protect your traffic. Moreover in Netsukuku, there will be a complete cryptographic layer, which guarantees security and anonymity for any connection (see Carciofo ).
The Netsukuku radar sends echo packets about every 10 seconds, if someone replies it communicates with it.
Yes, Viphilama will provide an automatic tunnelling system.
Whatever you already do in the current Internet, that is any IP application is supported.
ANDNA comes with a DNS wrapper so it is transparent to all the programs using glibc. Read the andna manual .
What do you mean by performance?
|it is dependent on the links quality. If the nodes are linked by 100Mbps cable you will feel like in a large LAN. The distance from yourself and the destination node is also relevant. Remember that the Netsukuku daemon chooses only the best way to reach the other nodes, but cannot improve the roads themselves.|
|Netsukuku has been designed to run on Access Point and low-spec devices. PyNtk uses a stripped version of python (under 500Kb).|
For now it runs only on GNU/Linux, but it is easy to port it on other OS. If you want to join in the development let us know ;)
For the Python implementation, you need iproute2 and stackless python. Stackless python normally isn't bundled with common GNU/Linux distributions, however a HOWTO is available in the Netsukuku site.
|Short answer:||if you code the port, yes.|
|Answer:||We need coders for that. There are a lot of things to be done and the Windows port is what we care less.|
We are currently working on flashing Netsukuku on Access Points (like Linksys). See http://netsukuku.freaknet.org/openwrt/
You just need a wifi network card. Put it in ad-hoc mode using "netsukuku" as essid. See http://netsukuku.freaknet.org/doc/manuals/ntk-wifi
Are you kidding?