Large scale distributed computing is a necessity for many companies nowadays. The huge data centers of these companies have to process large amounts of data, but also need to deal with constantly failing nodes. Fortunately, dealing with failing and joining nodes is a problem solved ages ago. However, what if you would push this flexible architectural paradigm of joining and leaving to the limit? In this project the student is asked to create a system JavaScript based distributed computing system that runs in the browser. The separate browser (and NodeJS) clients should be managed through one, or a series of master nodes and should be able to make computations for them in a reliable and secure way. The student will implement this system and demonstrate the workings with a simple task that is distributed over a network of failing and joining nodes.
Large scale distributed computing is a necessity for many companies nowadays. The huge data centers of these companies have to process large amounts of data, but also need to deal with constantly failing nodes. Fortunately, dealing with failing and joining nodes is a problem solved ages ago. However, what if you would push this flexible architectural paradigm of joining and leaving to the limit? In this project the student is asked to create a system JavaScript based distributed computing system that runs in the browser. The separate browser (and NodeJS) clients should be managed through one, or a series of master nodes and should be able to make computations for them in a reliable and secure way. The student will implement this system and demonstrate the workings with a simple task that is distributed over a network of failing and joining nodes.
Technologies: