Networking Basics: Physical Layer Transmission

Client-server model

By allowing clients to access remote services by simply calling a local procedure.

It automatically locates the correct server and sets up the communication (binding).

It can fragment and reassemble messages as needed, especially for large arrays.

Due to the RPC system's ability to hide details and handle data type conversions.

Simplified programming for most programmers.

The physical layer transmits the message by putting it onto the physical transmission medium.

The message is passed upward with each layer stripping off and examining its own header before arriving at the receiver, process Q.

A collection of protocols used in a particular system.

Middleware protocols are applications that logically reside in the OSI application layer.

DNS is a distributed service used to look up a network address associated with a name.

OSI protocols were not as popular due to the higher adoption of Internet protocols like TCP and IP.

Four combinations are: (a) Both sender and receiver executing, (b) Only sender executing, (c) Sender passive and receiver executing, (d) Both sender and receiver passive.

The receiver is passive in Figure 4.26(b), meaning it is in a state where message delivery is not possible.

In Figure 4.26(c), the sender is passive while the receiver is executing, allowing the receiver to read messages without the senders needing to be active.

Figure 4.26(d) shows the system storing and possibly transmitting messages even when both sender and receiver are passive.

The key requirement is that messages are properly addressed with a systemwide unique name of the destination queue.

It ensures that messages are not lost and can be retrieved at a later time, regardless of sender or receiver activity.

The major issue is that overlay networks need to be manually administrated.

The Channel Control Function allows an operator to monitor end points of a channel and manage queue managers hosting message channel agents.

In 2006, a working group formed to create the Advanced Message Queuing Protocol (AMQP).

AMQP evolves around applications, queue managers, and queues.

The implementation involves a single queue manager running on a separate server, with applications communicating through a local interface.

The purpose was to allow organizations to have more diverse messaging solutions beyond what a single manufacturer offers.

Flow control is established through credits, where a receiver informs the sender of the number of messages it can send over a specific link.

The three steps are: 1. Message assignment and recording in unsettled state, 2. Handling by the receiving application and reporting back, 3. Settling the message transfer between sender and receiver.

The three types are producer, consumer, and queue.

The sender assigns a unique identifier to the message and records it as unsettled before transferring it to the server.

After the receiver reports back that it has finished handling the message, the sender enters the settled state and informs the receiver to forget about the message.

Flow control, through credits, helps regulate the message exchange rate between sender and receiver, preventing overload and ensuring smooth communication.