𝑩𝒆𝒍𝒍’𝒔 𝑻𝒂𝒙𝒐𝒏𝒐𝒎𝒚 𝒇𝒐𝒓 𝑴𝑰𝑴𝑫 𝑨𝒓𝒄𝒉𝒊𝒕𝒆𝒄𝒕𝒖𝒓𝒆

According to Bell’s Taxonomy, the classification of computer architecture is primarily done on memory hierarchy. Going further in the Bell’s Taxonomy for MIMD Computers,  the MIMD(Multiple Instruction, Multiple Data) architecture is classified as MultiComputer and MultiProcessor. In the Multi Computer architecture, each computer has its own processor, local memory and I/O peripherals. The local memories are private to each computer and are not accessible to other processors. Message Passing interface is used to establish the communication between different processors in this type of architecture. In the MultiProcessor architecture, there is a global shared memory that can be accessed by all the processors.

Distributed Multicomputers: Distributed multicomputers are multicomputer systems where the individual computers (nodes) are geographically present across different locations. Each node operates as an independent computer with its own local memory and processors. Communication between distributed nodes is achieved through message passing over a network. Commonly used in applications that require collaboration and resource sharing across multiple locations

Central Multicomputers: Central multicomputers, also known as shared-memory multicomputers, are multicomputer systems in which multiple computers (nodes) share a central memory. Each node has its own processors, but they all access the same shared memory. Communication between nodes in central multicomputers occurs through shared memory operations, without the need for explicit message passing. Central multicomputers provide a shared-memory programming model and are suitable for applications that benefit from shared data access and synchronization

Distributed Multiprocessors: Distributed memory multiprocessors, also known as cluster-based multiprocessors, are multiprocessor systems where each processor has its own private memory. Processors communicate and exchange data by explicitly passing messages between each other over a network. They are designed to scale to a large number of processors and are commonly used in high-performance computing (HPC) applications that require parallel processing on large datasets.

Central Multiprocessors: Central memory multiprocessors, also referred to as shared-memory multiprocessors, are multiprocessor systems in which multiple processors share a single, global memory. All processors can directly access and modify shared data, enabling efficient communication and synchronization. Central memory multiprocessors typically provide a shared-memory programming model and are well-suited for applications that require high data sharing and inter-process communication.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
An Article by: Yashwanth Naidu Tikkisetty
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~