Introduction to Peer-to-Peer Networking

Peer to peer is an approach to computer networking where all computers share equivalent responsibility for processing data. Peer-to-peer networking (also known simply as peer networking) differs from client-server networking, where certain devices have responsibility for providing or "serving" data and other devices consume or otherwise act as "clients" of those servers.

Characteristics of Peer-to-Peer Network

- A P2P network can be an ad hoc connection—a couple of computers connected via a Universal Serial Bus to transfer files.
- A P2P network also can be a permanent infrastructure that links a half-dozen computers in a small office over copper wires.
- A P2P network can be a network on a much grander scale in which special protocols and applications set up direct relationships among users over the Internet.
- P2P software systems like Kazaa and Napster rank amongst the most popular software applications ever.
- P2P technologies promise to radically change the future of networking. P2P file sharing software has also created much controversy over legality and "fair use."
- A P2P network implements search and data transfer protocols above the Internet Protocol (IP).
- To access a P2P network, users simply download and install a suitable P2P client application.

A P2P setup can facilitate the following:
- Sharing of file by all the users of the network.
- Telephony.
- Media streaming - both audio and video.
- Community/discussion forums.

Types of P2P Networks

The P2P networks are normally classified as either ‘pure’ or ‘hybrid’ types.
Pure Networks
- Peers have got to be equals – no single node can supersede or dictate terms over another.
- There is no requirement of a central server and as such, no possibilty of a client-server correlation.
- There is also no need for a central router.

Hybrid P2P Networks
- This type of network needs to have a central server that can store data on all the peers and deliver it whenever asked to do so.
- The route terminals are treated as addresses, each one of which can be referenced by a specfic set of indices.
- Since the central server is not supposed to have any kind of resources available to it, the peers themselves are required to host all the resources. As and when required, a peer informs the central server about the type of resource to be shared and the details of the peer/s who should be allowed to share it.

Benefits of P2P Networks

1. Efficient use of resources.
- Unused bandwidth, storage, processing power at the edge of the network.
2. Scalability
- Consumers of resources also donate resources.
- Aggregate resources grow naturally with utilization.
3. Reliability
- Replicas.
- Geographic distribution.
- No single point of failure.
4. Ease of administration
- Nodes self organize.
- No need to deploy servers to satisfy demand (c.f. scalability).
- Built-in fault tolerance, replication, and load balancing.

Overview of LocalTalk Protocol

LocalTalk refers to the physical networking -- that means the built-in controller in many Apple computers, the cables and the expansion cards required on some systems. The "official" Apple cabling system typically uses a "bus topology" where each device in the network is directly connected to the next device in a daisy chain. The illustration on the Farallon book cover below gives an idea of how a bus looks.
Ethernet is the most-used method of Macintosh networking and all new Macs sport an Ethernet port, but the longevity of Macs mean there's still a bunch out there with serial ports (i.e., LocalTalk support) but no Ethernet. New iMacs and G4s lack serial ports, so can't network directly with older LocalTalk Macs and printers.

LocalTalk implementation utilized the Mac's RS-422 printer port with twisted-pair cabling and 3-pin DIN connectors. Systems were daisy-chained together and required adapters to work with the Mac's onboard DB-9 or 8-pin DIN connectors. LocalTalk provided a fairly speedy 230.4 kbps networking speed, very useable for file sizes and traffic levels of the day - compare this to the still used 56 kbps modem.

A variation of LocalTalk, called PhoneNet used standard unshielded twisted pair telephone wire with 6 position modular connectors (same as used in the popular RJ11 telephone connectors) connected to a PhoneNet transceiver, instead of the expensive shielded twisted-pair cable. In addition to being lower cost, PhoneNet-wired networks were more reliable due to the connections being more difficult to accidentally disconnect.
A LocalTalk-to-Ethernet Bridge is a network bridge that joins the physical layer of the AppleTalk networking used by previous generations of Apple Computer products to an Ethernet network. Some LocalTalk-to-Ethernet Bridges only performed Appletalk bridging. Others were also able to bridge other protocols. For example: TCP/IP in the form of MacIP.

Network Bridging

Bridging is a forwarding technique used in packet-switched computer networks. Unlike routing, bridging makes no assumptions about where in a network a particular address is located. Instead, it depends on flooding and examination of source addresses in received packet headers to locate unknown devices. Once a device has been located, its location is recorded in a table where the MAC address is stored so as to preclude the need for further broadcasting. The utility of bridging is limited by its dependence on flooding, and is thus only used in local area networks.
A network bridge is a device which connects two parts of a network together at the data link layer (layer 2 of the OSI model).

Situations Where Bridging Is Appropriate :
- Connecting Networks.
- Filtering/Traffic Shaping Firewall.
- Network Tap.
- Layer 2 VPN : Two Ethernet networks can be joined across an IP link by bridging the networks to an EtherIP tunnel or a tap(4) based solution such as OpenVPN.
- Layer 2 Redundancy : A network can be connected together with multiple links and use the Spanning Tree Protocol to block redundant paths.

Advantages :
- Self-configuring.
- Isolate collision domain.
- Reduce the size of collision domain by micro segmentation in non-switched networks.
- Transparent to protocols above the MAC layer.
- LANs interconnected are separate, and physical constraints such as number of stations, repeaters and segment length don't apply.
- Helps minimize bandwidth usage.

Disadvantages :
- It does not limit the scope of broadcasts.
- It does not scale to extremely large networks.
- Buffering and processing introduces delays.
- Bridges are more expensive than repeaters or hubs.

Network Interface Cards (NIC)

A Network Interface Card (NIC) provides the hardware interface between a computer and a network. A NIC technically is network adapter hardware in the form factor of an add-in card. Networked computers communicate with each other using a given protocol or agreed-upon language for transmitting data packets between the different machines, known as nodes. The network interface card acts as the liaison for the machine to both send and receive data on the LAN.
Some NIC cards work with wired connections while others are wireless. Most NICs support either wired Ethernet or WiFi wireless standards. In new computers, many NICs are now pre-installed by the manufacturer. NICs can be differentiated by their type of connectivity to the computer itself.

- 10/100 Ethernet : These cards are networking cards that are used most frequently in the home or small office setting. They are capable of speeds up to 10 or 100 megabits per second.
- Gigabit Ethernet :These NICs provide network transfer speeds of up to one Gigabit per second.
- Fiber Optics : These NICs use fiber optic cabling to reach speeds of 10 gigabits per second currently, with a specification under review to push this limit to 100 gigabits per second.
- Wireless NICs : These NICs provide the same networking capabilities as their wired counterparts, however, they have their own transfer capabilities. Speeds of 54 Mb/s are the most commonly available to wireless NICs without teaming several NICs together to combine bandwidths.
- Wireless Dongles : There is a wireless networking device used by individual machines that have access to a main computer that is connected to a wireless router. This wireless router allows the user to install wireless dongles instead of entire routers with each additional machine on the network.

Hybrid networks use a combination of any two or more topologies in such a way that the resulting network does not exhibit one of the standard topologies (e.g., bus, star, ring, etc.).
A standard hybrid network uses something called a hybrid access point, a networking device that both broadcasts a wireless signal and contains wired access ports. The most common hybrid access point is a hybrid router. The typical hybrid router broadcasts a Wi-Fi signal using 802.11 a, b or g and contains four Ethernet ports for connecting wired devices. The hybrid router also has a port for connecting to a cable or DSL modem via Ethernet cable.
There are several different possible network configurations for a hybrid network. The most basic configuration has all the wired devices plugged into the Ethernet ports of the hybrid router. Then the wireless devices communicate with the wired devices via the wireless router.
Computers aren't the only devices that can be linked over a hybrid network. You can now buy both wired and wireless peripheral devices like printers, Web cams and fax machines.

Hybrid network are: star ring network and star bus network
* A Star ring network consists of two or more star topologies connected using a multi-station access unit (MAU) as a centralized hub.
* A Star Bus network consists of two or more star topologies connected using a bus trunk (the bus trunk serves as the network's backbone).

While grid networks have found popularity in high-performance computing applications, some systems have used genetic algorithms to design custom networks that have the fewest possible hops in between different nodes. Some of the resulting layouts are nearly incomprehensible, although they function quite well.

Packet Sniffing

Packet sniffing is listening (with software) to the raw network device for packets that interest you. When your software sees a packet that fits certain criteria, it logs it to a file. The most common criteria for an interesting packet is one that contains words like "login" or "password."
Packet sniffing is a form of wire-tap applied to computer networks instead of phone networks. It came into vogue with Ethernet, which is known as a "shared medium" network. This means that traffic on a segment passes by all hosts attached to that segment. Ethernet cards have a filter that prevents the host machine from seeing traffic addressed to other stations. Sniffing programs turn off the filter, and thus see everyone's traffic.
A packet sniffer, sometimes referred to as a network monitor or network analyzer, can be used legitimately by a network or system administrator to monitor and troubleshoot network traffic. Using the information captured by the packet sniffer an administrator can identify erroneous packets and use the data to pinpoint bottlenecks and help maintain efficient network data transmission.
The versatility of packet sniffers means they can be used to:
* Analyze network problems.
* Detect network intrusion attempts.
* Gain information for effecting a network intrusion.
* Monitor network usage.
* Gather and report network statistics.
* Filter suspect content from network traffic.
* Spy on other network users and collect sensitive information such as passwords .
* Reverse engineer proprietary protocols used over the network.
* Debug client/server communications.
* Debug network protocol implementations.

Overview Of The Application Layer

Computer networks are inherently insecure. To keep information secret, it must be encrypted. Encryption protocols fall into two general classes: secret key (e.g. DES, IDEA), and public key (e.g. RSA). Using these protocols is straight-forward; the hard part is key management.
In addition to providing secrecy, cryptographic protocols can also provide authentication. Finally, cryptography can also be used to allow messages to be signed in such a way that the sender cannot repudiate them after they have been sent. Naming in the Internet uses a distributed database system, DNS. Domain Name Server(DNS) holds records with IP addresses, mail exchanges, and other information. By querying a DNS server, a process can map an Internet domain name onto the IP address used to communicate with that domain.
As networks grow larger, they become harder to manage. For this reason, special network management systems and protocols have been devised, the most popular of which is SNMP. This protocol allows managers to communicate with agents inside devices to read out their status and issue commands to them.
Four major network applications are electronic mail, USENET news, the World Wide Web, and multimedia. Most email systems use the mail system defined in RFCs 821 and 822. Messages sent in this system use system ASCII headers to define message properties. These messages are sent using SMTP. Two systems for securing email exist, PGP and PEM.
USENET news consists of thousands of newsgroups on all manner of topics. People can join newsgroups locally, and can then post messages all over the world using the NNTP protocol, which has some resemblance to SMTP.
The World Wide Web is a system for linking up hypertext documents. Each document is a page written in HTML, possible with hyperlinks to other documents. A browser can display a document by establishing a TCP connection to its server, asking for the document, and then closing the connection. When a hyperlink is selected by the user, that document can also be fetched in the same way. In this manner, documents all over the world are linked together in a giant web.
Multimedia is the rising star in the networking firmament. It allows audio and video to be digitized and transported electronically for display. Most multimedia projects use the MPEG standards and transmit the data over ATM connections.

Overview Of The Transport Layer

The transport layer is the key to understanding layered protocols. It provides various services, the most important of which is an end-to-end, reliable, connection-oriented byte stream from sender to receiver. It is accessed through service primitives that permit the establishment, use and release of connection.
Transport protocols must be able to do connection management over unreliable networks. Connection establishment is complicated by the existence of delayed duplicate packets that can reappear at inopportune moments. To deal with them, three-way handshakes are needed to establish connections. Releasing a connection is easier than establishing one, but is still far from trivial due to the two-army problem.
Even when the network layer is completely reliable, the transport layer has plenty of work to do. It must handle all the service primitives, manage connections and timers, and allocate and utilize credits.
The main Internet transport protocol is TCP. It uses a 20-byte header on all segments. Segments can be fragmented by routers within the Internet, so hosts must be prepared to do reassembly. A great deal of work has gone into optimizing TCP performance, using algorithms from Nagle, Clark, Jacobson, Karn and others.
ATM has four protocols in the AAL layer. All of them break messages into cells at the source and reassemble the cells into messages at the destination. The CS and SAR sublayers add their own headers and trailers in various ways, leaving from 44 to 48 bytes of cell payload.
Network performance is typically dominated by protocol and TPDU processing overhead, and the situation gets worse at higher speeds. Protocols should be designed to minimize the number of TPDUs, context switches, and times each TPDU is copied. For gigabit networks, simple protocols using rate, rather than credit, flow control are called for.

Overview Of The Medium Access Sublayer

Some networks have a single channel that is used for all communication. In these networks, the key design issue is the allocation of this channel among the competing stations wishing to use it. Numerous channel allocation algorithms have been devised like :

- FDM : Dedicate a frequency band to each station.
- TDM : Dedicate a time slot to each station.
- Pure ALOHA : Unsynchronized transmission at any instant.
- Slotted ALOHA : Random transmission in well defined time slots.
- 1-persistent CSMA : Standard carrier sense multiple access.
- Nonpersistent CSMA : Random delay when channel is sensed busy.
- P-persistent CSMA : CSMA, but with probability of p of persisting.
- CSMA/CD : CSMA. but abort on detecting a collision.
- Bit Map : Round robin scheduling using a bit map.
- Binary countdown : Highest numbered ready station goes next.
- Tree walk : Reduced contention by selective enabling.
- Wavelength division : A dynamic FDM scheme for fiber.
- MACA, MACAW : Wireless LAN protocols.
- GSM : FDM plus TDM for cellular radio.
- CDPD : Packet radio within an AMPS channel.
- CDMA : Everybody speak at once but in different language.
- Ethernet : CSMA/CD with binary exponential backoff.
- Token bus : Logical ring on a physical bus.
- Token Ring : Capture the token to send a frame.
- DQDB : Distributed queuing on a two-bus MAN.
- FDDI : Fiber-optic token ring.
- HIPPI : Crossbar using 50-100 twisted pairs.
- Fibre channel : Crossbar using fiber optics.
- SPADE : FDM with dynamic channel allocation.
- ACTS : TDM with centralized slot allocation.
- Binder : TDM with ALOHA when slot owner is not interested.
- Crowther : ALOHA with slot owner getting to keep it.
- Roberts : Channel time reserved in advance by ALOHA.

FDM and TDM are efficient when the number of stations is small and the traffic is continous.
ALOHA protocol, with and without slotting and control, has been proposed when the number of stations is large and variable.
BINARY COUNTDOWN completely eliminates contention.
TREE WALK reduces contention by dynamically dividing the stations into two disjoint groups, one of which is permitted to transmit and one of which is not.
WIRELESS LANs have their own problems and solutions. The biggest problem is caused by hidden stations, so CSMA does not work. MACA attempts to stimulate transmissions around the destination, to make CSMA work better.
GSM, CDPD and CDMA are widely used for mobile computers and telephones.
The IEEE 802 LANs are : CSMA/CD, TOKEN BUS, and TOKEN RING. Each of these has its unique advantages and disadvantages, and each has found its own user community.
FDDI, FAST ETHERNET, HIPPI, and FIBER CHANNEL offer bandwidth in the 100 Mbps range and up.

Quick Tech Lesson: Overview Of The Physical Layer

The physical layer is the basis of all the networks. Nature imposes two fundamental limits on all channels, and these determine their bandwidth. These limits are the Nyquist limit, which deals with noiseless channels, and the Shammon limit, for noisy channels.
Transmission media can be guided or unguided. The principle guided media are twisted pair, coaxial cable, and fiber optics. Unguided media include radio, microwaves, infrared, and lasers through the air.
A key element in most wide area networks is the telephone system. Its main components are the local loops, trunks and switches. Local loops are analog-twisted pair circuits, which require modems for transmitting digital data. Trunks are digital, and can be multiplexed in several ways, including FDM, TDM and WDM. The switches include crossbars, space divison switches, and time divison switches. Both circuit switching and packet switching are important.
The telephone system are now becoming digital from end to end and carrying both voice and data traffic over the same lines. Two variants of this new system, known as ISDN, are being introduced. Narrowband ISDN is a circuit switched digital system that is an incremental improvement over the current system. In contrast, broadband ISDN represents a paradigm shift, since it is based on cell switching ATM technology. Various kinds of ATM switches exist, including the knockout switch and the Batcher-banyan switch.
For mobile applications, the hard-wired telephone system is not suitable. Alternatives to the telephone system include cellular radio and communication satellites. Cellular radio is now widely used for portable telephones but will soon be common for data traffic as well. The current generation of cellular systems (e.g. AMPS) are analog, but the next generation (e.g. PCS/PCN) will be fully digital. Transmission communication satellites are geosynchronous, and there was some interest in bigger systems, but the failure of Iridium slowed down this process.