A study on application container resource efficiency

Abstract

Nowadays, the IT service environment develops in a dynamic, rapid, and unpredictable way. Microservices and application containers in this process have a significant impact on new generation IT service models. The fact that they have important capabilities such as modelability, presentability as service, and restructurability, are reasons for preferring them in many areas. Moreover, microservices can meet various needs of IT personnel. As it is known, all server system components, such as CPU, network, hard-drive I/O, affect energy consumption. At this point, microservices also play an important mediator role in resource management. Energy consumption of microservice-based applications is lower than that of the traditional approaches. However, there are still cases of recoverable and unnecessary consumption at some points. Microservices can be monitored and controlled using many methods. Thus, this provides us with opportunities to recover the wasted energy resources considerably. In this article, the effects of container-based microservice architectures on the energy consumption of the system and how to reduce these effects are presented. For this purpose, a methodology, which has 3 approaches (disconnect, pause, stop), and a tracing mechanism are proposed. The results show that this methodology has a considerable effect on energy efficiency.

Be Good Neighbors: A Novel Application Isolation Metric Used to Optimize the Initial Container Placement in CaaS

“…As a more flexible and efficient approach for resource sharing, CaaS offers easier manipulation for upper-layer applications than IaaS and more flexible support for the bottom-layer system than PaaS. Since containers are lighter than VMs and allow for high-density deployment, fine-grained resource management to reduce the servers' energy consumption has attracted increasing attention [7]- [9].…”