Distributed Memory Management

In the last two decades, the industry has seen a paradigm shift to big data and machine learning architectures that involve processing terabytes/petabytes of data as quickly as possible. As computing power became cheaper, there was a need to use multiple processors to speed up processing to a larger scale. This has led to distributed computing. Distributed computing refers to an arrangement of computer systems that are connected via some networking/distribution middleware. All the connected systems share resources and coordinate their activities via middleware so that they work in a way that is perceived as a single system by the end user. Distributed computing is needed due to the huge volume and throughput requirements of modern applications. Some typical examples of scenarios where computing demands cannot be fulfilled by a single system and that...