基于微服务架构FPGA云平台的并发请求调度机制

doi:10.19678/j.issn.1000-3428.0062678

计算机工程 ›› 2022, Vol. 48 ›› Issue (7): 206-213. doi: 10.19678/j.issn.1000-3428.0062678

基于微服务架构FPGA云平台的并发请求调度机制

奚智雯¹, 蔡晶晶¹, 阳文敏², 柴志雷^1,3

1. 江南大学人工智能与计算机学院, 江苏无锡 214122;
2. 江苏虎甲虫计算技术有限公司, 江苏无锡 214000;
3. 江苏省模式识别与计算智能工程实验室, 江苏无锡 214122

收稿日期:2021-09-14 修回日期:2021-10-15 出版日期:2022-07-15 发布日期:2021-10-19
作者简介:奚智雯(1997—),女,硕士研究生,主研方向为计算机体系结构;蔡晶晶,硕士研究生;阳文敏,硕士;柴志雷,教授、博士。
基金资助:
国家自然科学基金（61972180）。

Concurrent Request Scheduling Mechanism for FPGA Cloud Platform Based on Microservice Architecture

XI Zhiwen¹, CAI Jingjing¹, YANG Wenmin², CHAI Zhilei^1,3

1. School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu 214122, China;
2. Jiangsu Tiger Beetle Computing Technology Co., Ltd., Wuxi, Jiangsu 214000, China;
3. Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence, Wuxi, Jiangsu 214122, China

Received:2021-09-14 Revised:2021-10-15 Online:2022-07-15 Published:2021-10-19

摘要/Abstract

摘要： 基于微服务架构的现场可编程门阵列（FPGA）云平台在被大规模推广后积累了许多用户。针对FPGA云平台存在大量用户并发请求的问题，建立一种基于优先级调度的自定义参数响应指数计算模型。将5个请求关键影响因素作为自定义参数，采用层次分析法确定各参数权重，根据响应指数函数计算各请求的响应指数。在该模型的基础上，设计一种高并发请求调度（HCRS）算法，通过响应指数阈值对请求类别进行划分，使得高优先级请求优先得到处理，次优先级请求加入先进先出队列等待，低优先级请求暂时挂起，从而缩短请求响应时间以及请求响应延时，缓解由高并发请求带来的硬件节点资源分配压力。在真实运营的FPGA云平台中实现该算法并在实际环境中进行测试，结果表明，在并发请求量相同时，与先来先服务调度算法相比，HCRS算法的平均响应延时降低29 074 ms，平均请求响应时间缩短12 605 ms，其在提升系统吞吐量与并发度的同时可以有效优化硬件节点资源利用率。

关键词: 微服务架构, 现场可编程门阵列, 并发请求, 服务质量, 云计算, 优先级调度

Abstract: The Field Programmable Gate Array(FPGA) cloud platform based on microservices architecture has accumulated many users following its large-scale rollout.It has a problem with a large number of concurrent user requests.To address this problem, this paper establishes a calculation model of user-defined parameter response index based on priority scheduling, determines five request key influencing factors as custom parameters, uses the Analytic Hierarchy Process(AHP) to determine the weight of each parameter, and calculates the response index of each request according to the response index function.Then, a High Concurrent Request Scheduling(HCRS) algorithm is designed based on this model.This algorithm classifies requests using a response index threshold.Therefore, the highest priority requests are processed first, secondary priority requests are added to the first-in-first-out queue to wait, and low priority requests are temporarily suspended.This method shortens the request response time and latency, easing the hardware node resource allocation pressure caused by high concurrent requests.Experiments compared the performance of the HCRS algorithm with the First Come, First Serve(FCFS) algorithm on the FPGA cloud platform in a real scenario.The results show that average response latency of the HCRS algorithm decreased by 29 074 ms, and the average request response time decreased by 12 605 ms.The proposed contribution improves throughput, system concurrency, and utilization of hardware node resources.

Key words: microservice architecture, Field Programmable Gate Array(FPGA), concurrent request, Quality of Service(QoS), cloud computing, priority scheduling

中图分类号:

TP391

奚智雯, 蔡晶晶, 阳文敏, 柴志雷. 基于微服务架构FPGA云平台的并发请求调度机制[J]. 计算机工程, 2022, 48(7): 206-213.

XI Zhiwen, CAI Jingjing, YANG Wenmin, CHAI Zhilei. Concurrent Request Scheduling Mechanism for FPGA Cloud Platform Based on Microservice Architecture[J]. Computer Engineering, 2022, 48(7): 206-213.

https://www.ecice06.com/CN/Y2022/V48/I7/206

图/表 17

20221029174933

20221029174936

20221029174939

20221029174943

20221029174946

20221029174950

20221029174954

20221029174958

20221029175001

20221029175005

20221029175009

20221029175013

20221029175017

20221029175021

20221029175024

20221029175029

20221029175033

参考文献

[1] ALVES G R, FIDALGO A, MARQUES A, et al.Spreading remote lab usage a system-a community-a Federation[C]//Proceedings of the 2nd International Conference of the Portuguese Society for Engineering Education.Washington D.C., USA:IEEE Press, 2016:1-7.
[2] 李康, 张鲁飞, 张新伟, 等.基于FPGA集群的脉冲神经网络仿真器设计[J].计算机工程, 2020, 46(10):201-209. LI K, ZHANG L F, ZHANG X W, et al.Design of spiking neural network simulator based on FPGA cluster[J]. Computer Engineering, 2020, 46(10):201-209.(in Chinese)
[3] DOSHI J, PATIL P, DAVE Z, et al.Implementing a cloud based Xilinx ISE FPGA design platform for integrated remote labs[C]//Proceedings of International Conference on Advances in Computing, Communications and Informatics.Washington D.C., USA:IEEE Press, 2015:533-537.
[4] FUJII N, KOIKE N.IoT remote group experiments in the cyber laboratory:a FPGA-based remote laboratory in the hybrid cloud[C]//Proceedings of International Conference on Cyberworlds.Washington D.C., USA:IEEE Press, 2017:162-165.
[5] TOYODA Y, KOIKE N, LI Y M.An FPGA-based remote laboratory:implementing semi-automatic experiments in the hybrid cloud[C]//Proceedings of the 13th International Conference on Remote Engineering and Virtual Instrumenta-tion.Washington D.C., USA:IEEE Press, 2016:24-29.
[6] 施凌鹏, 朱征, 周俊松, 等.面向微服务架构的云系统负载均衡机制[J].计算机工程, 2021, 47(9):44-50, 58. SHI L P, ZHU Z, ZHOU J S, et al.Load balancing mechanism for microservice architecture in cloud-based systems[J].Computer Engineering, 2021, 47(9):44-50, 58.(in Chinese)
[7] FOWLER M, LEWIS J.Microservices[EB/OL].[2021-07-05].https://martinfowler.com.
[8] KAINZ A.Microservices vs.Monoliths:an operational comparison[EB/OL].[2021-07-05].https://thenewstack.io/microservices-vs-monoliths-an-operational-comparison.
[9] LIU X, ZHAI X B, LU W D, et al.QoS-guarantee resource allocation for multibeam satellite industrial Internet of Things with NOMA[J].IEEE Transactions on Industrial Informatics, 2021, 17(3):2052-2061.
[10] SHETTY J, ANTONY D'MELLO D.QoS based cloud service selection to handle large volume of concurrent requests[J].Indian Journal of Science and Technology, 2016, 9(48):15-26.
[11] SUN M Y, ZHOU Z B, WANG J P, et al.Energy-efficient IoT service composition for concurrent timed applications[J].Future Generation Computer Systems, 2019, 100:1017-1030.
[12] 郭军, 武静, 邢留冬, 等.面向突发业务的云服务并发量应对策略研究[J].计算机学报, 2019, 42(4):864-882. GUO J, WU J, XING L D, et al.A coping strategy for bursty workload of cloud service[J].Chinese Journal of Computers, 2019, 42(4):864-882.(in Chinese)
[13] AMAL MURAYKI ALRUWAILI B A A, HUMAYUN M, JHANJHI N Z.Proposing a load balancing algorithm for cloud computing applications[J].Journal of Physics:Conference Series, 2021, 1979(1):012034.
[14] LIU W, ZHANG D W, GAO Z J.A novel load balancing strategy based on node load comprehensive measuring under cloud computing environment[J].Journal of Computational Methods in Sciences and Engineering, 2019, 19:3-9.
[15] ZHOU J, CEN B W, CAI Z X, et al.Workload modeling for microservice-based edge computing in power Internet of Things[J].IEEE Access, 2021, 9:76205-76212.
[16] AL-DEBAGY O, MARTINEK P.A comparative review of microservices and monolithic architectures[C]//Proceedings of the 18th International Symposium on Computational Intelligence and Informatics.Washington D.C., USA:IEEE Press, 2018:149-154.
[17] VILLAMIZAR M, GARCÉS O, OCHOA L, et al.Cost comparison of running Web applications in the cloud using monolithic, microservice, and AWS Lambda architectures[J].Service Oriented Computing and Applications, 2017, 11(2):233-247.
[18] LU X Z, PHANG K.An enhanced Hadoop heartbeat mechanism for MapReduce task scheduler using dynamic calibration[J].China Communications, 2018, 15(11):93-110.
[19] ROCHIM A F, AZIZ M A, FAUZI A.Design log management system of computer network devices infrastructures based on ELK stack[C]//Proceedings of International Conference on Electrical Engineering and Computer Science.Washington D.C., USA:IEEE Press, 2019:338-342.
[20] 邓雪, 李家铭, 曾浩健, 等.层次分析法权重计算方法分析及其应用研究[J].数学的实践与认识, 2012, 42(7):93-100. DENG X, LI J M, ZENG H J, et al.Research on computation methods of AHP wight vector and its applications[J].Mathematics in Practice and Theory, 2012, 42(7):93-100.(in Chinese)
[21] 刘卫锋, 何霞.基于模糊判断矩阵的两阶段群决策方法[J].计算机工程, 2012, 38(10):141-143. LIU W F, HE X.Two-step method of group decision making based on fuzzy judgment matrix[J].Computer Engineering, 2012, 38(10):141-143.(in Chinese)
[22] RUDRABHATLA C K.A quantitative approach for estimating the scaling thresholds and step policies in a distributed microservice architecture[J].IEEE Access, 2020, 8:180246-180254.
[23] BAE J, JANG H, JIN W J, et al.Jointly optimizing task granularity and concurrency for in-memory MapReduce frameworks[C]//Proceedings of IEEE International Conference on Big Data.Washington D.C., USA:IEEE Press, 2017:130-140.

选择文件类型/文献管理软件名称

选择包含的内容

基于微服务架构FPGA云平台的并发请求调度机制

Concurrent Request Scheduling Mechanism for FPGA Cloud Platform Based on Microservice Architecture

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	关明晓, 刘嘉堃, 张鸿锐, 何安平. 基于FPGA误差可控的浮点运算加速器研究[J]. 计算机工程, 2024, 50(5): 291-297.
[2]	杨思捷, 陈俊奇, 王勇, 李树林. 基于FPGA的软硬件协同纠删码编码加速方案[J]. 计算机工程, 2024, 50(2): 224-231.
[3]	孙毅, 王会梅, 鲜明, 向航. Kubeflow异构算力调度策略研究[J]. 计算机工程, 2024, 50(2): 25-32.
[4]	李浩阳, 贺小伟, 王宾, 吴昊, 尤琪. 基于改进Informer的云计算资源负载预测[J]. 计算机工程, 2024, 50(2): 43-50.
[5]	余长宏, 陆雅, 王海鑫, 高明. 基于滑动时间窗的物联网设备流量分类算法[J]. 计算机工程, 2023, 49(7): 259-268.
[6]	刘志彬, 黄秋兰, 胡庆宝, 程耀东, 胡誉, 田浩来. Kubernetes异构资源细粒度调度策略的设计与实现[J]. 计算机工程, 2023, 49(2): 31-36,45.
[7]	陈逸, 刘博生, 徐永祺, 武继刚. 混合精度频域卷积神经网络FPGA加速器设计[J]. 计算机工程, 2023, 49(12): 1-9.
[8]	洪起润, 王琴. 基于帧间数据复用的稀疏CNN加速器设计[J]. 计算机工程, 2023, 49(12): 55-62.
[9]	王恩旭, 王晓红, 张坤, 张冬雯. 基于双重注意力机制的云计算负载预测模型[J]. 计算机工程, 2023, 49(11): 40-48, 69.
[10]	黄正伟, 刘宏伟, 徐渊. 用于IToF传感器的极低功耗RISC-V专用处理器设计[J]. 计算机工程, 2022, 48(9): 146-154.
[11]	黄金瑶, 刘同来, 吴嘉鑫, 武继刚. 多周期家庭护理的路径规划与调度算法[J]. 计算机工程, 2022, 48(7): 292-299.
[12]	张垿豪, 冯文龙, 黄梦醒, 刘伟. 基于区块链的科技服务质量信任评价方案[J]. 计算机工程, 2022, 48(5): 127-135,144.
[13]	王奎宇, 宋晓勤, 缪娟娟, 张昕婷, 雷磊. 基于SDN的高性能QoS保障低轨道卫星星间路由算法[J]. 计算机工程, 2022, 48(5): 185-190,199.
[14]	贺小伟, 徐靖杰, 王宾, 吴昊, 张博文. 基于GRU-LSTM组合模型的云计算资源负载预测研究[J]. 计算机工程, 2022, 48(5): 11-17,34.
[15]	巩杰, 赵烁, 何虎, 邓宁. 基于FPGA的量化CNN加速系统设计[J]. 计算机工程, 2022, 48(3): 170-174,196.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于微服务架构FPGA云平台的并发请求调度机制

Concurrent Request Scheduling Mechanism for FPGA Cloud Platform Based on Microservice Architecture

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献

相关文章 15

编辑推荐

Metrics

本文评价