Live Migration of Multi-Container Kubernetes Pods in Multi-Cluster Serverless Edge Systems

被引:0
|
作者
Poggiani, Leonardo [1 ]
Puliafito, Carlo [1 ]
Virdis, Antonio [1 ]
Mingozzi, Enzo [1 ,2 ]
机构
[1] Univ Pisa, Dept Informat Engn, Pisa, Italy
[2] Univ Florence, Dept Informat Engn, Florence, Italy
来源
PROCEEDINGS OF THE 2024 WORKSHOP ON SERVERLESS AT THE EDGE, SEATED 2024 | 2024年
关键词
Serverless; Function as a Service; Edge computing; Multi-cluster; Kubernetes; Migration;
D O I
10.1145/3660319.3660330
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
In the cloud-native landscape, serverless computing is establishing itself as a pillar for smart provisioning of modern microservice-based applications. Due to its resounding success, serverless is rapidly emerging as a powerful paradigm in edge computing as well, wherein computing resources are geographically distributed in proximity to users. In serverless edge systems, applications can be composed by two types of microservice instances. Remote-state instances are internally stateless and access a remote state, if needed. On the other hand, a local-state instance is dedicated to a user and locally retains state information. In this work, we focus on live migration of local-state instances across an edge system operated by Kubernetes, which is the de-facto standard for microservice automated orchestration. More specifically, we propose a solution able to migrate both single-container and multi-container instances between different Kubernetes clusters. Our approach leverages the Liqo open-source project to establish a peering relationship and transfer the instance state between the involved clusters. After presenting the design and proof-of-concept implementation of the proposed solution, we outline useful insights into how to deploy it in practice and describe the experiments carried out to validate our work. Our system represents a powerful Kubernetes extension that does not require any modification to its standard API.
引用
收藏
页码:9 / 16
页数:8
相关论文
共 50 条
  • [21] Deep learning-based edge caching for multi-cluster heterogeneous networks
    Jiachen Yang
    Jipeng Zhang
    Chaofan Ma
    Huihui Wang
    Juping Zhang
    Gan Zheng
    Neural Computing and Applications, 2020, 32 : 15317 - 15328
  • [22] Deep learning-based edge caching for multi-cluster heterogeneous networks
    Yang, Jiachen
    Zhang, Jipeng
    Ma, Chaofan
    Wang, Huihui
    Zhang, Juping
    Zheng, Gan
    NEURAL COMPUTING & APPLICATIONS, 2020, 32 (19): : 15317 - 15328
  • [23] Multi-cluster MIMO non-orthogonal multiple access for multi-cell systems
    Shin, Changyong
    WIRELESS NETWORKS, 2024, 30 (04) : 2187 - 2201
  • [24] Multi-cluster visualization and live reporting of Static Analysis Security Testing (SAST) warnings
    Pathak, Abhishek
    Sivakumar, Kaarthik
    Haque, Mazhar
    Ganesan, Prasanna
    2019 IEEE SECURE DEVELOPMENT (SECDEV 2019), 2019, : 145 - 145
  • [25] Coordinated Power Allocation for Generalized Multi-Cluster Distributed Antenna Systems
    Feng, Wei
    Wang, Yanmin
    Li, Yunzhou
    Xu, Xibin
    Wang, Jing
    IEICE TRANSACTIONS ON COMMUNICATIONS, 2011, E94B (09) : 2656 - 2659
  • [26] Schedulability analysis and optimisation for the synthesis of multi-cluster distributed embedded systems
    Pop, P
    Eles, P
    Peng, Z
    IEE PROCEEDINGS-COMPUTERS AND DIGITAL TECHNIQUES, 2003, 150 (05): : 303 - 312
  • [27] Localized communications of data parallel programs on multi-cluster grid systems
    Hsu, CH
    Lo, TT
    Yu, KM
    ADVANCES IN GRID COMPUTING - EGC 2005, 2005, 3470 : 900 - 910
  • [28] Schedulability analysis and optimization for the synthesis of multi-cluster distributed embedded systems
    Pop, P
    Eles, P
    Peng, Z
    DESIGN, AUTOMATION AND TEST IN EUROPE CONFERENCE AND EXHIBITION, PROCEEDINGS, 2003, : 184 - 189
  • [29] Network-aware selective job checkpoint and migration to enhance co-allocation in multi-cluster systems
    Jones, William M.
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2009, 21 (13): : 1672 - 1691
  • [30] Multi-Container Migration Strategy Optimization for Industrial Robotics Workflow Based on Hybrid Tabu-Evolutionary Algorithm
    Xie, Xingju
    Wu, Xiaojun
    Hu, Qiao
    Yuan, Sheng
    IEEE TRANSACTIONS ON SERVICES COMPUTING, 2024, 17 (05) : 2640 - 2653