基于日志可观测的云—边协同微服务部署优化

doi:10.11959/j.issn.1000-0801.2023254

摘要/Abstract

摘要：

依托微服务、容器化等云原生能力，数字化业务可拆分成细粒度的微服务模块，通过云边协同的方式优化微服务部署模式。一般来说，边缘计算可通过将计算资源扩展到网络边缘侧，降低物联网应用的响应时延及运行开销，但边缘侧设备计算及存储能力有限，通常难以满足组合服务多类型资源需求。因此，将中心云与边缘侧设备的微服务功能进行划分，是提高服务质量、优化网络性能的必要手段。为了实现微服务模块的合理配置，提出了一种基于日志可观测的云—边协同微服务部署优化方法，构建基于微服务的时序过程模型，并通过时序约束挖掘算法发现微服务间连接关系。采用多目标优化算法，最小化微服务配置时延和能耗，实现云—边协同微服务部署优化。最后，实验验证了方法的有效性。

关键词: 服务部署, 云边协同, 云原生可观测性, 时序约束, 多目标优化

Abstract:

Relying on cloud native capabilities including micro-services and containerization, digital services can be partitioned into fine-grained micro-service modules, and micro-service deployment strategy can be optimized through cloud-edge collaboration.Generally, edge computing reduces network response latency and running overhead through expanding computing resources to the edge of network.However, edge devices have limited computing and storage capacity, which is usually difficult to fulfill types of resource requirements for composed micro-services.Therefore, achieving micro-service functionalities division on the cloud and edge devices is a necessary means to improve quality of service and optimize network performance.A service configuration optimization approach for cloud-edge collaboration based on log observability was proposed to complete the reasonable configuration of micro-service modules.A timed micro-service-based process model was constructed, and connected relations between micro-services were discovered by a temporal constraints mining algorithm.A multi-objective optimization algorithm was adopted to minimum service delay and energy consumption for optimizing the micro-service deployment in the cloud-edge network.The effectiveness of the approach was verified by experiments.

Key words: service deployment, cloud-edge collaboration, cloud-native observability, temporal constraints, multi-objective optimization

中图分类号:

TP393

孙梦宇, 林睿. 基于日志可观测的云—边协同微服务部署优化[J]. 电信科学, 2023, 39(12): 122-132.

Mengyu SUN, Rui LIN. Cloud-edge collaboration service deployment optimization based on log observability[J]. Telecommunications Science, 2023, 39(12): 122-132.

图/表 10

图1

图2

图3

表1

图4

图5

图6

图7

图8

图9

参考文献 14

[1]	WANG L , JIAO L , LI J ,et al. MOERA:mobility-agnostic online resource allocation for edge computing[J]. IEEE Transactions on Mobile Computing, 2019,18(8): 1843-1856.
[2]	LIU X Z , SUN S X , HUANG G . Decentralized services computing paradigm for blockchain-based data governance:programmability,interoperability,and intelligence[J]. IEEE Transactions on Services Computing, 2020,13(2): 343-355.
[3]	SAHNI J , VIDYARTHI D P . A cost-effective deadline- constrained dynamic scheduling algorithm for scientific workflows in a cloud environment[J]. IEEE Transactions on Cloud Computing, 2018,6(1): 2-18.
[4]	CHEN X , SHI Q , YANG L ,et al. ThriftyEdge:resource- efficient edge computing for intelligent IoT applications[J]. IEEE Network, 2018,32(1): 61-65.
[5]	SONG A , CHEN W N , LUO X N ,et al. Scheduling workflows with composite tasks:a nested particle swarm optimization approach[J]. IEEE Transactions on Services Computing, 2022,15(2): 1074-1088.
[6]	CHEN L X , ZHOU P , GAO L ,et al. Adaptive fog configuration for the industrial Internet of Things[J]. IEEE Transactions on Industrial Informatics, 2018,14(10): 4656-4664.
[7]	LI R , ZHOU Z , CHEN X ,et al. Resource price-aware offloading for edge-cloud collaboration:a two-timescale online control approach[J]. IEEE Transactions on Cloud Computing, 2022,10(1): 648-661.
[8]	FADAHUNSI O , MAHESWARAN M . Locality sensitive request distribution for fog and cloud servers[J]. Service Oriented Computing and Applications, 2019,13(2): 127-140.
[9]	LI X P , YU W , RUIZ R ,et al. Energy-aware cloud workflow applications scheduling with geo-distributed data[J]. IEEE Transactions on Services Computing, 2022,15(2): 891-903.
[10]	LYU X C , TIAN H , SENGUL C ,et al. Multiuser joint task offloading and resource optimization in proximate clouds[J]. IEEE Transactions on Vehicular Technology, 2017,66(4): 3435-3447.
[11]	CUI L Z , XU C , YANG S ,et al. Joint optimization of energy consumption and latency in mobile edge computing for Internet of things[J]. IEEE Internet of Things Journal, 2019,6(3): 4791-4803.
[12]	MANNHARDT F , DE LEONI M , REIJERS H A ,et al. Datadriven process discovery-revealing conditional infrequent behavior from event logs[C]// International Conference on Advanced Information Systems Engineering. Cham:Springer, 2017: 545-560.
[13]	ZHOU Z B , CHENG Z H , ZHANG L J ,et al. Scientific workflow clustering and recommendation leveraging layer hierarchical analysis[J]. IEEE Transactions on Services Computing, 2018,11(1): 169-183.
[14]	LERA I , GUERRERO C , JUIZ C . Availability-aware service placement policy in fog computing based on graph partitions[J]. IEEE Internet of Things Journal, 2019,6(2): 3641-3651.

参数	设置值
边缘侧设备微服务容量C	2
微服务所需CPU周期数μ_k(m)/MB	[50, 200]
边缘侧设备CPU频率f_n/GHz	4
中心云CPU频率f₀/GHz	8
微服务任务量τ_k(m)/MB	[10, 100]
边缘侧设备间传输带宽w_n/（Mbit·s^-1）	1
边缘侧设备与中心云间传输带宽w₀/（Mbit·s^-1）	1
边缘侧设备与中心云传输固有时延t₀/s	2
边缘侧设备静态能耗E₀/J	1
数据包传输和接收的能耗常数E_elec/（J·bit^-1）	5×10^-4
发送电子常数ε_amp/（J·bit^-1·m^-2）	1×10^-6
传输衰减指数p	2

[1]	朱姝. 云网融合时代基于边缘云的智能云化家庭网关的探索与实践[J]. 电信科学, 2023, 39(Z1): 1-6.
[2]	张剑, 程俊华, 龚菡洁, 李红, 牛凯. 基于云边协同的高可用厨房卫生监控系统[J]. 电信科学, 2023, 39(Z1): 62-70.
[3]	朱姝. 边缘云网下的新业务探索与应用[J]. 电信科学, 2023, 39(Z1): 55-61.
[4]	薛飞, 陈彬, 刘静, 梁晓扬, 朱琳, 王凤, 李天, 张靓, 陈贞贞, 李潇. 自智网络全栈部署技术研究与实践[J]. 电信科学, 2023, 39(9): 63-75.
[5]	沈林江, 曹畅, 崔超, 张岩. 基于策略约束强化学习的算网多目标优化研究[J]. 电信科学, 2023, 39(8): 136-148.
[6]	王淑玲, 孙杰, 王鹏, 杨爱东. 云边协同中的资源调度优化[J]. 电信科学, 2023, 39(2): 163-170.
[7]	王文哲, 安岗, 吴健, 赵文东, 王延刚, 赵云峰, 狄子翔, 顾照杰, 邹贵祥. 一种支持云原生的工业互联网业务管理平台[J]. 电信科学, 2022, 38(Z1): 203-213.
[8]	朱洪涛, 郭庆. 基于用户群组的低轨卫星网络多星切换策略[J]. 电信科学, 2022, 38(4): 39-48.
[9]	陈瑞凤, 李君, 徐春婕, 程清波, 吕晓军. 5G在铁路智能车站中的应用[J]. 电信科学, 2021, 37(7): 86-95.
[10]	唐博恒, 柴鑫刚. 基于云边协同的计算机视觉推理机制[J]. 电信科学, 2021, 37(5): 72-81.
[11]	武宇亭, 王旭亮, 全硕. KubeTelecom：一种面向5G网络切片的多云多容器集群管理与运维引擎[J]. 电信科学, 2021, 37(12): 72-83.
[12]	崔羽飞,张云勇,张第,王晓霞,潘思宇,李堃. 基于云边协同的工业互联网实践[J]. 电信科学, 2020, 36(2): 109-116.
[13]	高永梅,程冠杰. 基于边缘计算的数据密集型服务部署[J]. 电信科学, 2019, 35(7): 87-99.
[14]	包晓安,曹云棣,张娜,钱俊彦,曹建文. 基于格分布方差的多目标云工作流调度算法[J]. 电信科学, 2019, 35(2): 1-13.
[15]	张煜培,赵知劲,郑仕链. 利用多目标PSO优化的累积时延和信道容量联合优化的频谱切换算法[J]. 电信科学, 2018, 34(7): 62-71.