A contribution to scheduling jobs submitted by finite-sources in computational clusters

Data science and data processing are very popular topics nowadays. Un- like a few years ago, everything is connected to now we have handle these kinds of large well. Therefore the distributed heterogeneous resources networks e.g. computational grid, attracted great interest. It has become challenge schedule jobs in order utilize available effectively. The allocation arriving impact on efficiency energy consumption system. A generalized finite source model presented this paper. Our main goal build up models for performance evaluation scheduling computeintensive with unknown service times cluster that consists servers different types. For purpose determine various measures all combinations three policies (two them novelty paper: MRT MRTHP policies) which can be used assigning schemes buffering jobs. Furthermore, investigate effect switching off idle system under schemes. Computational results obtained by simulation show choice policy scheme plays an important role ensuring quality parameters such as waiting time response experienced case However, only affected saving mode, while does not significant impact.