TIPC introduction 1

TIPC Implemented as a kernel module , Exists in all major Linux Distribution in . It can be configured to pass UDP Or transmit messages directly through Ethernet . Message passing is sequence guarantee 、 No loss and flow control . The delay time is shorter than any other known Protocol , The maximum throughput is the same as TCP Quite a .

Have you ever wished that even in cluster It can also be used when transmitting data between nodes Unix Domain Socket ？ You decide the location of the address you want to bind and use ？ You do not have to perform DNS Find and worry IP Address ？ You don't have to start a timer to monitor the continuous existence of peer sockets ？ But without the disadvantages of this socket type , Like delay inode The risk of ？
Welcome to the transparent interprocess communication service , In short ,TIPC, It provides you with all these , And more .

Service Addressing（ Service addressing ）
A fundamental concept in TIPC is that of Service Addressing which makes it possible for a programmer to chose his own address, bind it to a server socket and let client programs use only that address for sending messages.
TIPC A basic concept in is service addressing , It allows programmers to choose their own addresses , Bind it to the server socket , And let the client program send messages only with this address .

Service Tracking（ Service tracking ）
A client wanting to wait for the availablity of a server, uses the Service Tracking mechanism to subscribe for binding and unbinding/close events for sockets with the associated service address.
The service tracking mechanims can also be used for Cluster Topology Tracking, i.e., subscribing for availability/non-availability of cluster nodes.
Likewise, the service tracking mechanims can be used for Cluster Connectivity Tracking, i.e., subscribing for up/down events for individual links between cluster nodes.
Clients who want to wait for server availability use the service tracking mechanism to subscribe to bind and unbind sockets with associated service addresses / Closing event .
The service tracking mechanism can also be used for cluster topology tracking , That is, the availability of subscription cluster nodes / Unavailability .
similarly , The service tracking mechanism can be used for cluster connectivity tracking , namely , Subscribe up for a single link between cluster nodes / Downward event .

Transmission Modes（ transmission mode ）
Using a service address, a client can send datagram messages to a server socket.
Using the same address type, it can establish a connection towards an accepting server socket.
It can also use a service address to create and join a Communication Group, which is the TIPC manifestation of a brokerless message bus.
Multicast with very good performance and scalability is available both in datagram mode and in communication group mode.
Use the service address , The client can send datagrams to the server socket （datagram） news .
Use the same address type , It can establish a connection to the socket of the receiving server .
It can also create and join communication groups using service addresses , This is agentless message bus TIPC form .
In datagram mode and communication group mode , Multicast has good performance and scalability .

Inter Node Links（ Link between nodes ）
Communication between any two nodes in a cluster is maintained by one or two Inter Node Links, which both guarantee data traffic integrity and monitor the peer node’s availability.
cluster The communication between any two nodes in is maintained by one or two inter node links , These two links ensure the integrity of data communication , The availability of peer nodes is also monitored .

Cluster Scalability（Cluster Extensibility ）
By applying the Overlapping Ring Monitoring algorithm on the inter node links it is possible to scale TIPC clusters up to 1000 nodes with a maintained neighbor failure discovery time of 1-2 seconds. For smaller clusters this time can be made much shorter.
By applying the overlapping ring monitoring algorithm on the link between nodes （Overlapping Ring Moniroring）, Can be TIPC The cluster extends to 1000 Nodes , Keep the neighbor fault discovery time as 1-2 second . For smaller clusters , This time can be shortened a lot .

Neighbor Discovery
Neighbor Node Discovery in the cluster is done by Ethernet broadcast or UDP multicast, when any of those services are available. If not, configured peer IP addresses can be used.
When any of these services is available , The neighbor nodes in the cluster are discovered through Ethernet broadcast or UDP Multicast to complete . without , Configured peers can be used IP Address .

Configuration
Run in single node mode TIPC when , No configuration required . When running in cluster mode ,TIPC You must obtain at least one node address （ stay linux4.17 Before ）, And tell which interface to connect to .“tipc” Configuration tools make it possible to add and maintain more configuration parameters .

Performance（ performance ）
TIPC Message transmission latency is better than any other known Protocol . The maximum byte throughput of connections between nodes is still slightly lower than TCP, On the same host , They are better than TCP.