What is a buffer and why is it used in ICM?
A. A buffer is a holding place for data. The processes in ICM use or can use buffering. Buffering is used when a process is unable to transmit data to another process at a specific time. For example, on a Peripheral Gateway (PG), the Peripheral Gateway Agent (PGAG) process communicates to the router using the Central Controller Agent (CCAG) process. When the PGAG process on a specific PG is unable to communicate to the CCAG process, it stores the information intended to send into a buffer. The most common causes for the use of a buffer are network-related problems such as latency or complete loss of connections. When such performance-related issues cause the use of buffers, if the issues are not corrected, the buffers continue to fill. Buffers have a fixed size. If a buffer reaches its limit, the associated process has no way of sending the information to its peer process and clear its own buffer. Eventually, the process that is using the buffer fails because it is overrun with data. Y
