Title :
Workload Modeling and its Implications on Data-Center-on-a-Chip Optimization: From Mathematical Models to Control Algorithms
Author_Institution :
Ming Hsieh Dept. of Electr. Eng., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
The design and optimization of data-centers-on-a-chip (DCoC) requires accurate mathematical modeling of the workloads. We describe a statistical physics-inspired formalism that allows both the derivation of state space representation of the network and the dynamic optimization.
Keywords :
computer centres; dynamic programming; network theory (graphs); network-on-chip; DCoC; control algorithms; data-center-on-a-chip optimization; dynamic optimization; mathematical models; network representation; state space representation; statistical physics-inspired formalism; workload modeling; Analytical models; Fractals; Mathematical model; Network-on-chip; Optimal control; Optimization; Routing; Networks-on-Chip; fractality; fractional calculus; fractional optimal control; nonstationarity; workloads;
Conference_Titel :
Control Systems and Computer Science (CSCS), 2015 20th International Conference on
Conference_Location :
Bucharest
Print_ISBN :
978-1-4799-1779-2
DOI :
10.1109/CSCS.2015.142
