Method of QoS evaluation of FPGA as a service

Artem Perepelitsyn, Vitaliy Kulanov, Inna Zarizenko

Abstract


The subject of study in this article is the evaluation of the performance issues of cloud services implemented using FPGA technology. The goal is to improve the performance of cloud services built on top of multiple FPGA platforms known as FPGA-as-a-Service (FaaS). Task: to analyze the delays in communications between host computer and FPGA; propose the steps of development to reduce the delay and perform the evaluation of the response time for the FPGA-based accelerator depending on number of involved cards; consider the reliability aspect of such systems implemented using programmable logic. According to the tasks, the following results were obtained. The FPGA-as-a-Service where FPGA resources are provided through a set of hardware/software toolset is considered. The usage of queueing theory for cloud-based services is analyzed. The contribution of the parts of FPGA-as-a-Service to the final delay of the service is discussed. The process of modeling of work the services based on FPGA accelerator cards with use of Jackson's network is analyzed in detail. The model of the delays of FaaS that considers the parameters of accelerator FPGA cards is offered. The formula of the total response time of the service combined based on the response of the components of is obtained. The proposed steps of reduce data processing delays include increase the size of data blocks for processing in FPGA by each kernel, change the communication model with kernel from sequential to pipelined, following timing closure technique and use more FPGA accelerator cards in parallel to divide the enquiring delay. Based on the proposed model the evaluation of response time of FaaS was done. The advantage of the use of many FPGAs in parallel for same data processing task instead of implementation of requests thread for each accelerator card is shown. Conclusions. The main contribution of this study is a step forward to the modeling of FPGA-based services that can be used for FPGA-based artificial intelligence (AI) applications. It helps to improve the performance of the system by means of reducing the delays at different stages of requests processing. Another side of this result is the reliability aspect that is based on modified manner of service operation in case of use the proposed steps of system optimization. It helps to improve the processing of requests to FaaS. The proposed method is the next step after prototyping of such systems because it helps to turn the FaaS from the object for development to the tool for deployment of new technologies like AI applications.

Keywords


FPGA; FPGA-as-a-Service; FaaS; cloud infrastructure; queueing theory; performance; reliability

Full Text:

PDF

References


Perepelitsyn, A., Zarizenko, I., Kulanov, V. FPGA as a Service Solutions Development Strategy. Proceedings 2020 IEEE 11th International Conference on Dependable Systems, Services and Technologies, DESSERT 2020, 2020, pp. 376-380. DOI: 10.1109/DESSERT50317.2020.9125017.

Zarizenko, I., Perepelitsyn, A. Analysis of tools and technologies of FaaS development. Radioelectronic and Computer Systems, 2019, no. 4, pp. 88-93. DOI: 10.32620/reks.2019.4.10.

Di Mauro, M., Liotta, A., Longo, M., Postiglione, F. Statistical Characterization of Containerized IP Multimedia Subsystem through Queueing Networks. Proceedings of 2020 6th IEEE Conference on Network Softwarization, NetSoft 2020, 2020, pp. 100–105. DOI: 10.1109/NetSoft48620.2020.9165357.

Kulanov, V., Perepelitsyn, A., Zarizenko, I. Method of development and deployment of reconfigurable FPGA-based projects in cloud infrastructure. Proceedings of 2018 IEEE 9th International Conference on Dependable Systems, Services and Technologies, DESSERT 2018, 2018, pp. 103-106. DOI: 10.1109/DESSERT.2018.8409108.

Containerizing Alveo Accelerated Applications with Docker. Available: https://xilinx.com/developer/articles/containerizing-alveo-accelerated-application-with-docker.html. (accessed January 27, 2020).

Tsimashenka, I., Knottenbelt, W. J. Reduction of Subtask Dispersion in Fork-Join Systems. Computer Performance Engineering. EPEW 2013. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, 2013, vol. 8168, pp. 325–336. DOI: 10.1007/978-3-642-40725-3_25.

Samuilov, K., Zaryadov, I., Gorbunova, A. Analysis of the response time of a cloud computing system. IX International industry scientific and technical conference "Information Society Technologies", 2015. pp. 29–30.

Gaidamaka, Yu., Sopin, E., Talanova, M. A Simplified model for performance analysis of cloud computing systems with dynamic scaling. Proc. of the 18th International Scientific Conference "Distributed Computer and Communication Networks: Control, Computation, Communications", DCCN 2015, 2015, pp. 75–86.

Vats, S., Kumar Sharma, S., Kumar, S. A Switch Based Resource Management Method for Energy Optimization in Cloud Data Center. International Journal of Computing, 2021, vol. 20, iss. 1, pp. 85–91. DOI: 10.47839/ijc.20.1.2103.

Gad-Elrab, A. A. A., Alzohairy, T. A., Raslan, K. R., Emara, F. A. Genetic-Based Task Scheduling Algorithm with Dynamic Virtual Machine Generation in Cloud Computing. International Journal of Computing, 2021. vol. 20, iss. 2, pp. 165–174. DOI: 10.47839/ijc.20.2.2163.

Sai Sowjanya, T., Praveen, D., Satish, K., Rahiman, A. The queueing theory in cloud computing to reduce the waiting time. International Journal of Computer Science Engineering and Technology, 2011, vol. 1, iss. 3, pp. 110–112.

Alveo U280 Data Center Accelerator Card User Guide, Xilinx, UG1314 (v1.3). Available at: https://www.sandycast.com/support/documentation/boards_and_kits/accelerator-cards/ug1314-u280-reconfig-accel.pdf. (accessed February 27, 2020).

XRT Controlled Kernel Execution Models. Available: https://xilinx.github.io/XRT/master/html/xrt_kernel_executions.html. (accessed October 7, 2022).

Vitis Unified Software Platform Documentation: Application Acceleration Development, Xilinx, UG1393 (v2019.2). Available at: https://docs.xilinx.com/r/en-US/ug1393-vitis-application-acceleration. (accessed February 28, 2020).

UltraFast Design Methodology Guide for the Vivado Design Siute, Xilinx, UG949 (v2019.2). Available at: https://docs.xilinx.com/v/u/2019.2-English/ug949-vivado-design-methodology. (accessed December 6, 2019).

Martyniuk, T., Krukivskyi, B., Kupershtein, L., Lukichov, V. Neural network model of hetero-associative memory for the classification task. Radioelectronic and Computer Systems, 2022, no. 2, pp. 108–117. DOI: 10.32620/reks.2022.2.09.

Moskalenko, V., Moskalenko, A. Neural network based image classifier resilient to destructive perturbation influences – architecture and training method. Radioelectronic and Computer Systems, 2022, no. 3, pp. 95–109. DOI: 10.32620/reks.2022.3.07.




DOI: https://doi.org/10.32620/reks.2022.4.12

Refbacks

  • There are currently no refbacks.