Scheduling computation and communication on a software-defined photonic Network-on-Chip architecture for high-performance real-time systems

2018 ◽  
Vol 90 ◽  
pp. 54-71 ◽  
Author(s):  
Hüseyin Temuçin ◽  
Kayhan M. İmre
2018 ◽  
Vol 35 (5) ◽  
pp. 19-27 ◽  
Author(s):  
Adam Kostrzewa ◽  
Sebastian Tobuschat ◽  
Rolf Ernst

Nanophotonics ◽  
2018 ◽  
Vol 7 (5) ◽  
pp. 827-835 ◽  
Author(s):  
Hao Jia ◽  
Ting Zhou ◽  
Yunchou Zhao ◽  
Yuhao Xia ◽  
Jincheng Dai ◽  
...  

AbstractPhotonic network-on-chip for high-performance multi-core processors has attracted substantial interest in recent years as it offers a systematic method to meet the demand of large bandwidth, low latency and low power dissipation. In this paper we demonstrate a non-blocking six-port optical switch for cluster-mesh photonic network-on-chip. The architecture is constructed by substituting three optical switching units of typical Spanke-Benes network to optical waveguide crossings. Compared with Spanke-Benes network, the number of optical switching units is reduced by 20%, while the connectivity of routing path is maintained. By this way the footprint and power consumption can be reduced at the expense of sacrificing the network latency performance in some cases. The device is realized by 12 thermally tuned silicon Mach-Zehnder optical switching units. Its theoretical spectral responses are evaluated by establishing a numerical model. The experimental spectral responses are also characterized, which indicates that the optical signal-to-noise ratios of the optical switch are larger than 13.5 dB in the wavelength range from 1525 nm to 1565 nm. Data transmission experiment with the data rate of 32 Gbps is implemented for each optical link.


Author(s):  
Harendra Kumar ◽  
Isha Tyagi

Distributing tasks to processors in distributed real time systems is an important step for obtaining high performance. Scheduling algorithms play a vital role in achieving better performance and high throughput in heterogeneous distributed real time systems. To make the best use of the computational power available, it is essential to assign the tasks to the processor whose characteristics are most appropriate for the execution of the tasks in a distributed processing system. This study develops two algorithms for clustering the heavily-communicating tasks to reduce the inter-tasks communication costs by using k-means and fuzzy c-means clustering techniques respectively. In order to minimize the system cost and response time, an algorithm is developed for the proper allocation of formed clusters to the most suitable processor. The present algorithms are collated with problems in literature. The proposed algorithms are formulated and applied to numerous numerical examples to demonstrate their effectiveness.


Sign in / Sign up

Export Citation Format

Share Document