SMCFA: A Zynq-based stacked multi CPU-FPGA architecture

With an emerging market of ARM embedded processors and rapidly updating techniques used in manufacture process of FPGA, the structure of CPU combining with FPGA is increasingly attracting attention, such as the Zynq SoC heterogeneous computing platform developed by Xilinx which integrates a dual-core Cortex-A9 ARM processor and a FPGA in one chip. In this paper, we propose an 4-node Zynq based stacked multi CPU-FPGA architecture SMCFA. The dual-core ARM processors act as transaction process handler, and the FPGA implements the function of hardware acceleration. Four Zynq boards connected into stacked multi CPU-FPGA architecture by an external bus, which is specifically designed for this structural called SEU-mBUS. For experiment, we utilize SMCFA to accelerate satellite image retrieval. The Accelerator IP core is 9.2× faster than i5-3470 CPU when handling images, and using SMCFA can achieve a 5.3× speedup on the entire satellite image retrieval system.

[1]  Houman Homayoun,et al.  Accelerating Machine Learning Kernel in Hadoop Using FPGAs , 2015, 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing.

[2]  Yu Wang,et al.  FPMR: MapReduce framework on FPGA , 2010, FPGA '10.

[3]  Paul Chow,et al.  ZCluster: A Zynq-based Hadoop cluster , 2013, 2013 International Conference on Field-Programmable Technology (FPT).

[4]  Sanjay Ghemawat,et al.  MapReduce: Simplified Data Processing on Large Clusters , 2004, OSDI.

[5]  Stephen Neuendorffer,et al.  Building zynq® accelerators with Vivado® high level synthesis , 2013, FPGA '13.

[6]  Siddharth Garg,et al.  Energy-efficient computing from systems-on-chip to micro-server and data centers , 2015, 2015 Sixth International Green and Sustainable Computing Conference (IGSC).

[7]  Huazhong Yang,et al.  FPMR: MapReduce Framework on FPGA A Case Study of RankBoost Acceleration , 2010 .

[8]  Li Xiang,et al.  Ant Cluster: A Novel High-Efficiency Multipurpose Computing Platform , 2015 .

[9]  Sanjay Ghemawat,et al.  MapReduce: a flexible data processing tool , 2010, CACM.

[10]  Toshimori Honjo,et al.  Hardware acceleration of Hadoop MapReduce , 2013, 2013 IEEE International Conference on Big Data.

[11]  Nachiket Kapre,et al.  Zedwulf: Power-Performance Tradeoffs of a 32-Node Zynq SoC Cluster , 2015, 2015 IEEE 23rd Annual International Symposium on Field-Programmable Custom Computing Machines.