Department of Computer Engineering
Partitioning Models For Simultaneous Load Balancing on Computation and Data in Big Data Applications
Mestan Fırat Çeliktuğ
(Supervisor: Prof. Dr. Cevdet Aykanat)
Computer Engineering Department
There has been proposed several successful partitioning models for computational load balancing of irregularly sparse applications on distributed-memory architectures in the literature. However, there is a lack of partitioning models that encode both computational and data load balancing of processors. In our previous work, we proposed partitioning models simultaneously encoding computational and data load balancing of processors in graph and hypergraph applications. In recent years, tensor is a data structure which is of the focus in the literature of Big Data Processing. In the study, tensors are also planned to be focused in terms of the multi-weight balancing.
DATE: 02 December 2020, Wednesday @ 13:55