PROJECT TITLE :
Towards Max-Min Fair Resource Allocation for Stream Big Data Analytics in Shared Clouds - 2018
Distributed stream huge information analytics platforms have emerged to tackle the continuously generated knowledge streams. In stream massive knowledge analytics, the info processing workflow is abstracted as a directed graph referred to as a topology. Data are browse from the storage and processed tuple by tuple, and these processing results are updated dynamically. The performance of a topology is evaluated by its throughput. This Project proposes an economical resource allocation scheme for a heterogeneous stream huge data analytics cluster shared by multiple topologies, so as to achieve max-min fairness within the utilities of the throughput for all the topologies. We have a tendency to initial formulate a novel resource allocation problem, that is a mixed 0-one integer program. The NP-hardness of the matter is rigorously proven. To tackle this problem, we tend to remodel the non-convex constraint to many linear constraints using linearization and reformulation techniques. Based mostly on the analysis of the problem-specific structure and characteristics, we have a tendency to propose an approach that iteratively solves the continuous drawback with a fastened set of discrete variables optimally, and updates the discrete variables heuristically. Simulations show that our proposed resource allocation theme remarkably improves the max-min fairness in utilities of the topology throughput, and is low in computational complexity.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here