Hi all -- with this post I elaborate on the Transmission Control Protocol, including the the basics of initiating, using and ending a TCP connection.
TCP is used extensively for webpages (WWW), FTP, SSH, Email, and much else. It's one of the foundational protocols for internet communciations, so
]]>Hi all -- with this post I elaborate on the Transmission Control Protocol, including the the basics of initiating, using and ending a TCP connection.
TCP is used extensively for webpages (WWW), FTP, SSH, Email, and much else. It's one of the foundational protocols for internet communciations, so it's well worth understanding it's tradeoffs. It's optimized for accurate delivery over timely delivery, so this ought to be taken into consideration when assessing the right protocol for your transmission needs.
Data structure
TCP provides reliability in network packet transmission. Data is sent as a stream of 1 byte octets. Octet is stated explicitely to clarity that there is 8 bits of data, with no parity bit that is often found in storage bytes. Packets are sent using a duplex virtual circuit, which means that they can be sent simultaneously between both transmitting hosts.
The basic data unit in the TCP protocol is a "segment". Each segment contains a sequence number, acknowledgement number, window, checksum, source and destination port, along with additional metadata, the actual data, and padding.
How it works
All packets containing data are expected to be acknowledged as received. In the case where they are lost or corrupted, they are sent again until a timeout has occured. TCP requires a heavier implementation with a more complicated handshakes in the setup and teardown to enable a stable, reliable connection.
TCP provides basic operation signals for ensuring a successful TCP lifecycle, and they communicate a variety of situations:
URG: Urgent pointer field is valid
ACK: Acknowledgement field is valid
PSH: This segment requests a push
RST: Reset the connection
SYN: Synchronize sequence numbers
FIN: Sender has reached end of its byte stream
ACK segments do not require an acknowledgement, as this would incur an infinite loop.
Requesting a TCP connection
Establishing a virtual circuit uses a 3-way handshake, and the connection stays open until the end of the data transfer.
Closing a TCP connection
In theory, a TCP connection can stay open indefinitely. The close sequence is as follows:
Once the TCP connection has been closed, there is a cooldown period to ensure that there is no confusion between this connection and the next connection.
Reliable transmission and network congestion
In contrast to UDP, packets in TCP can't be fired and forgotten. The throughput of all data transmitted over TCP is managed, requiring an acknowledgement for each data packet sent within a window of acceptable data sequences allotted by the transmitting host.
A sequence number is assigned to each byte transmitted, and requires a positive aknowledgement from the receiving host. If the ACK is not received within a timeout period, the data is retransmitted with a longer timeout period until the max is hit.
The window shifts over the entirety of the data sequence until they have all been received. This window regulates TCP throughput, and also decreases network congestion by having hosts transmit only the data that the receiving end is ready to receive.
Thanks for reading. Feel free to tweet at me if you have any questions.
Part three will be a breakdown of the different major routing protocols:
Border Gateway Protocol (BGP), Routing Information Protocol (RIP), and Open Shortest Path First (OSPF).
Sources
]]>This is my first post in a series covering the basics of internet networking. This is a new topic to me, so I'll likely spend the next couple of weeks focused on learning this material. Feel free to tweet at me if you have any requests or suggestions.
]]>This is my first post in a series covering the basics of internet networking. This is a new topic to me, so I'll likely spend the next couple of weeks focused on learning this material. Feel free to tweet at me if you have any requests or suggestions.
Prehistory
In 1974 Vint Cert and Bob Kahn published a paper entitled "A Protocol for Packet Network Intercommunication", which described a way for sharing resources between different networks. The ideas and original "Transmission Control Program" model were refactored into a more modular system adhering to the end-to-end principle, known today as the Internet Protocol (IP) Suite.
Packet Switching Networks
The original packet-switching network was originally included in the early department of defence ARPANET by Donald Davies. It's well-known that the ARPANET is the direct anticedant to the internet, and it's because of this that packet-switching networks are the primary basis for data communications in computer networks worldwide.
In contrast to circuit-switching, which pre-allocates network resources for each communication, packet switching enables all channels to be agnostic of the data being sent through, so long as each bit of data has a distinguishable header explaining how it is to be treated. For instance, in a circuit-switching network, a communication session is leased at a certain bitrate and latency for a certain period of time, while a packet-switching network can handle transferring any type of data at any bitrate, so long as it conforms to the datagram structure, comprised of a header and paylaod.
Each channel in a packet-switching network can pass through any variable bit rate stream of packets, so long as the header information is available indicating the final destination. As the packets encounter switches and router, they are received, buffered, queued, and forwarded, immediately freeing up the channel for other packets.
In a connectionless packet-switching network, each packet header includes a destination address, source address, port numbers, and other metadata required for a successful routing. The packets are routed individually, and can arrive out of order. The packets are refered to as Datagrams (being the underlying data type of IP), and the process is refered to as datagram-switching. Example of protocols that use datagram-switching are Internet Protocol, User Datagram Protocol (which adds a few nice features such as sumchecking) and Ethernet protocol.
A connection-oriented network uses a virtual circuit (or virtual connection, or virtual channel,) which is established between all nodes between the origin and the destination. From that point, much of the header information can be stripped out, enabling a simple byte-stream of data to be delivered between the nodes. Crucially, a virtual circuit enables higher-level protocols to simply deal with the data without having to deal with individual digital tranmission units, such as segments, packets, or frames. The destination node receives the data in order.
Internet Protocol
The "IP" in TCP/IP refers to the fundamental connection layer: a connectionless datagram service responsible for relaying datagrams across a network in a "best-effort" fashion. It doesn't handle security, or integrity, or the order in which data is received -- it's only job is to route packets based on the IP addresses in the packet headers.
IP Protocols
IP protocols (how IP messages are formed, sent, and received) have developed over time. We're currently seeing a transition from IPv4 to IPv6.
IPv4 is the most widely used protocol version number is because the first three were experimental versions, used between 1977 and 1979. The successor to IPv4 is IPv6, and currently accounts for approximately 25% of internet addresses. The overwhelming reason to switch to IPv6 is that there are more of them available. Much, much more.
IPv4 uses 32 bits for addressing, which means there can only exist 4.3 billion addresses. When we consider the growth of the internet, and the inclusion of a whole "internet of things", it's obvious that this is not nearly enough. Ipv6 uses 128-bit addresses, which wikipedia tells me yields 340 undecillion addresses. That's a lot of of IOT lightbulbs.
Another major distinction between IPv4 and IPv6 is that IPv4 has the ability to automatically fragment an original datagram provided by IP into smaller units for transmission, while IPv6 does not, and relies on end stations and higher-level protocols for this, which is consistent with the end-to-end principle.
There are also differences in the class types of the addresses. IPv4 has five class types (A, B, C, D, E), with D reserved for multicast. Ipv6 has three multiple class types: Unicast, Multicast, and Anycast.
Finally, the representation of the addresses is quite different. Most people are familiar with the look of IPv4 address, in their common "xxx.xxx.xxx.xxx" format, with "localhost" as "127.0.0.1". IPv6 looks much more complicated, as it is denoted by eight groups of hex quartets, separated by collons: "xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx".
While the overriding reason to switch to IPv6 is the availability of more addresses, the additional differences in the protocol allow to more efficient routing and reduced management requirement between nodes on the network.
Transport Layers
On top of IP, at the transport layer, is found the more sophisticated features of communications networking. UDP retains the connectionless datagram approach of data transission, but with an optional checksum to provide data integrity. TCP provides a connection-oriented service allowing more flexibility and reliability in data transmission between two parties.
Data tranfered in TCP is called a "segment", and TCP uses sequence and acknowledge numbers to recover lost segments, detect out-of-order segments, and resolve transmission errors. This ensures a reliable stream of segments, with the added overhead of a more complicated handshake when setting up the virtual circuit.
My next post will cover the workings of this handshake, how TCP can provide secure connections, and how it provides for HTTP and Websockets.
]]>