Availability optimization of consistency and availability-based micro-service systems through elastic scheduling of container resources

The availability of consistency (C) and availability (A)-based micro-service systems is low when both consistency and partition tolerance (P) are satisfied. Considering the low resource occupation and fast supply of containers, this paper puts forward an approach to optimize the availability of CP micro-service systems based on the elastic scheduling of container resources, and sets up a prediction model of response time using the cascade queuing system. Then, the author determined whether to relax, restrict or maintain the container resource in light of the conformity of the response time. Finally, the proposed optimization approach was verified through experiments. The results show that a 2~3s-long adaptation period is needed for the approach under abrupt load changes, and the response time can be accurately predicted to ensure the system availability in the other cases. RÉSUMÉ. La disponibilité des systèmes de micro-service basés sur la cohérence (C) et la disponibilité (A) est faible lorsque la cohérence et la tolérance de partition (P) sont satisfaites. Compte tenu de la faible occupation des ressources et de l'approvisionnement rapide de conteneurs, cet article propose une approche visant à optimiser la disponibilité des systèmes de micro-service CP basés sur la planification élastique des ressources de conteneur et établit un modèle de prévision du temps de réponse à l'aide du système de mise en file d'attente en cascade. Ensuite, l’auteur a déterminé s’il fallait assouplir, restreindre ou maintenir la ressource du conteneur compte tenu de la conformité du temps de réponse. Enfin, l'approche d'optimisation proposée a été vérifiée par des expériences. Les résultats montrent qu’une période d’adaptation de 2 à 3 secondes est nécessaire pour l’approche lors de changements brusques de la charge et que le temps de réponse peut être prédit avec précision pour garantir la disponibilité du système dans les autres cas.

[1]  Kurt Rothermel,et al.  HAWKS: A System for Highly Available Executions of Workflows , 2016, 2016 IEEE International Conference on Services Computing (SCC).

[2]  Jian-Feng Cai,et al.  Two-phase approach for deblurring images corrupted by impulse plus gaussian noise , 2008 .

[3]  Ling Li.  Research on Data Consistency and Concurrency Optimization of Distributed System , 2012 .

[4]  Jian Lu,et al.  Probabilistically-Atomic 2-Atomicity: Enabling Almost Strong Consistency in Distributed Storage Systems , 2017, IEEE Transactions on Computers.

[5]  Xi Liu,et al.  > Replace This Line with Your Paper Identification Number (double-click Here to Edit) < , 2022 .

[6]  Carl Boettiger,et al.  An introduction to Docker for reproducible research , 2014, OPSR.

[7]  Sam Newman,et al.  Building Microservices , 2015 .

[8]  Wu Heng,et al.  Elastic Resource Provisioning Approach for Container in Micro-Service Architecture , 2017 .

[9]  Ahmad Habibizad Navin,et al.  CONSISTENCY OF DATA REPLICATION PROTOCOLS IN DATABASE SYSTEMS : A REVIEW , 2014 .

[10]  Vittaldas V. Prabhu,et al.  An integrated software system for enterprise performance management , 2007 .

[11]  Cao Guizhen,et al.  JMeter-based aging simulation of computing system , 2010, 2010 International Conference on Computer, Mechatronics, Control and Electronic Engineering.

[12]  Werner Vogels,et al.  Dynamo: amazon's highly available key-value store , 2007, SOSP.

[13]  Xiaoyun Zhu,et al.  Triage: performance isolation and differentiation for storage systems , 2004, Twelfth IEEE International Workshop on Quality of Service, 2004. IWQOS 2004..

[14]  Mazin S. Yousif,et al.  Microservices , 2016, IEEE Cloud Comput..

[15]  Ludmila Cherkasova,et al.  Session-Based Admission Control: A Mechanism for Peak Load Management of Commercial Web Sites , 2002, IEEE Trans. Computers.