Optimizing the Performance of a Microservice-Based Application Deployed on User-Provided Devices

Abstract

It is now feasible for consumers to buy inexpensive devices that can be installed at home and accessed remotely thanks to an Internet connection. Such a simple "self-hosting" can be an alternative to traditional cloud providers, especially for privacy-conscious users. We discuss how a community of users can pool their devices in order to host microservices-based applications, where each microservice is deployed on a different device. The performance of such an application depends heavily on the computing and network resources that are available and on the placement of each microservice. Finding the placement that minimizes the application response time is an NP-hard problem. We show that, thanks to well known optimization techniques (Particle Swarm Optimization), it is possible to quickly find a service placement resulting in a response time close to the optimal one. Thanks to an emulation platform, we evaluate the robustness of this solution to changes in the Quality of Service under conditions typical of a residential access network.

Swarm Intelligence for Next-Generation Wireless Networks: Recent Advances and Applications

“…The experiments using an Adult database verify the efficiency of the privacy-preserved model against information loss risks [157]. Moreover, the work in [158] presents a data privacy-preserved PSO algorithm for cloud computing. Especially, the traditional cloud is replaced by a cloud microservice that is deployed at the mobile device instead of remote cloud servers.…”

“…Enabled by the high optimization ability by simulating bird predation behavior, SI-based methods such as PSO can build security-preserved network models for 5G IoT networks to promote universal data exchange without compromising data privacy requirements [154]. Recently, some data privacy-preserved SI algorithms are also considered for SDN and cloud computing-based 5G networks [158] by forming service migration functions with privacy awareness. Other SI techniques such as ACO [160] also facilitate the 5G network authentication with bio-inspired algorithms embedded at the network controllers, e.g., SDN controllers, for securing communications and authenticating users with QoS preservation.…”

QoS-aware Autonomic Adaptation of Microservices Placement on Edge Devices

“…We characterized the impact of these variations of the network QoS on the response-time of the photosharing application by deploying its microservices on the same devices used for the above measurements according to the solution presented in (Stevant et al, 2018). The devices are dedicated to this experiment, therefore computing resources are always available for the application.…”

“…The algorithm used in the Plan function is based on the PSO meta-heuristic presented in (Stevant et al, 2018). A first input parameter for this algorithm is the call-graph of the microservices composing the application, as shown in Figure 1 for the case of the photosharing application.…”

“…The MetaHub (MH) service stores photo metadata. We have previously shown (Stevant et al, 2018) how the Particle Swarm Optimization (PSO) heuristic can place each microservice on user-provided devices to minimize the response time (weighted sum of response times corresponding to different user cases, to be precise), in a static case, i.e., where the different parameters (Round Trip Time (RTT), bandwidth, etc.) do not change over time.…”