Performance Optimization of Underlying Operating System in Transparent Computing

Transparent Computing: Development and Current Status

“…Xue et al [59] use stochastic Petri nets to model the performance of TC and provided mathematical analysis on the response latency, throughput, and wait time. In the following, we summarize the related works about the performance optimization of TC in Table 3 [60][61][62][63][64][65][66][67][68][69][70] .…”

“…Optimization type Performance LBTC [60] Load balance i) Reduce average response time ii) Improve system load capacity E-Eifel [61] NSAP i) Reduce the number of pseudo-timeouts ii) Improve the data transmission efficiency of NSAP ONSA and OFSA [62] Task scheduling Achieve 1.1 times acceleration than the optimal solution 2HR-fτ [63] Routing algorithm i) Improve the network throughput ii) Reduce deliver delay HG [64] File fetch and transmission Reduce total file fetch time nearly by 50% TCOS [65] Kernel and application layer i) Reduce the time of booting the OS by 33.11% ii) Decrease the memory space usage by 40.43% SDSCS [66] Code scheduling Improve application performance and user experience DCA [67] Resource allocation Improve the TC system performance TL [68] Intelligent TC system i) Integrate TC with machine learning ii) Reduce the training time considerably HIF [69] Task offloading Reduce the offloading energy TCC [70] Cache optimization i) Decrease the usage of server disk I/O ii) Improve startup speed of the terminal devices iii)Decrease the network bandwidth are interconnected via networks. In such a mixed computing architecture, the service demands of these devices are typically delay-sensitive and context-aware.…”