Hierarchical Scaling of Microservices in Kubernetes

Rossi, Fabiana; Cardellini, Valeria; Presti, Francesco Lo

doi:10.1109/acsos49614.2020.00023

Cited by 41 publications

(30 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are several proposals for resource allocation in the case of networks and cloud computing for achieving QoE [5], [11]- [26]. Broadly, we can classify these works as those involving computing (Kubernetes based) [21]- [25] and the ones involving networking (SDN based) [5], [11]- [20]. The slicing approaches formulated over SDN typically find/reconfigure routes and decide on bandwidth allocations along the routes.…”

Section: Background and Literature Reviewmentioning

confidence: 99%

“…This assumption is quite restrictive as applications or users would be only concerned about E2E QoE, and the mapping between E2E QoE and resources is often complex, dynamic, and stochastic. The authors in [21]- [25] mostly deal with Kubernetes settings and container deployment strategies. However, they do not incorporate the associated networking delays.…”

Section: Background and Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Achieving AI-enabled Robust End-to-End Quality of Experience over Radio Access Networks

Roy¹,

Rao²,

Alpcan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Emerging applications such as Augmented Reality, the Internet of Vehicles and Remote Surgery require both computing and networking functions working in harmony. The End-to-end (E2E) quality of experience (QoE) for these applications depends on the synchronous allocation of networking and computing resources. However, the relationship between the resources and the E2E QoE outcomes is typically stochastic and non-linear. In order to make efficient resource allocation decisions, it is essential to model these relationships. This article presents a novel machine-learning based approach to learn these relationships and concurrently orchestrate both resources for this purpose. The machine learning models further help make robust allocation decisions regarding stochastic variations and simplify robust optimization to a conventional constrained optimization. When resources are insufficient to accommodate all application requirements, our framework supports executing some of the applications with minimal degradation (graceful degradation) of E2E QoE. We also show how we can implement the learning and optimization methods in a distributed fashion by the Software-Defined Network (SDN) and Kubernetes technologies. Our results show that deep learning-based modelling achieves E2E QoE with approximately 99.8% accuracy, and our robust joint-optimization technique allocates resources efficiently when compared to existing differential services alternatives.

show abstract

Section: Background and Literature Reviewmentioning

confidence: 99%

Section: Background and Literature Reviewmentioning

confidence: 99%

Achieving AI-enabled Robust End-to-End Quality of Experience over Radio Access Networks

Roy¹,

Rao²,

Alpcan³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Compute Gradients Existing Slices: ∇π i (s i (k)); Estimate Gradient for New/Unknown Slice: ∇π j (s j (k)) using (10);…”

Section: Algorithm 1: Reconfiguration Algorithm With Online Gradient ...mentioning

confidence: 99%

“…At each step of the algorithm, the new slice is allocated more resources. To calculate (10), the SDN probes the network and server with ∆ increments around the current allocation point s j (k). Since the delays and throughput are stochastic, the probing is done multiple times and the relevant statistics (like average or max or any other percentile) depending on requirement is used to calculate (10).…”

Section: Algorithm 1: Reconfiguration Algorithm With Online Gradient ...mentioning

confidence: 99%

Online Slice Reconfiguration for End-to-End QoE in 6G Applications

Roy¹,

Rao²,

Alpcan³

et al. 2022

Preprint

View full text Add to dashboard Cite

End-to-end (E2E) quality of experience (QoE) for 6G applications depends on the synchronous allocation of networking and computing resources, also known as slicing. However, the relationship between the resources and the E2E QoE outcomes is typically stochastic and non-stationary. Existing works consider known resource demands for slicing and formulate optimization problems for slice reconfiguration. In this work, we create and manage slices by learning the relationship between E2E QoE and resources. We develop a gradient-based online slice reconfiguration algorithm (OSRA) to reconfigure and manage slices in resource-constrained scenarios for radio access networks (RAN). We observe that our methodology meets the QoE requirements with high accuracy compared to existing approaches. It improves upon the existing approaches by approximately 98% for bursty traffic variations. Our algorithm has fast convergence and achieves low E2E delay violations for lower priority slices.

show abstract

“…CPU utilization percentage represents the average CPU utilization of all copies of the current pod. A Pod's CPU utilization is the Pod's current CPU usage divided by the Pod Request value [14].…”

Section: Cpu Utilizationmentioning

confidence: 99%

A Fine-Grained Horizontal Scaling Method for Container-Based Cloud

Jiang

2021

Scientific Programming

View full text Add to dashboard Cite

The container scaling mechanism, or elastic scaling, means the cluster can be dynamically adjusted based on the workload. As a typical container orchestration tool in cloud computing, Horizontal Pod Autoscaler (HPA) automatically adjusts the number of pods in a replication controller, deployment, replication set, or stateful set based on observed CPU utilization. There are several concerns with the current HPA technology. The first concern is that it can easily lead to untimely scaling and insufficient scaling for burst traffic. The second is that the antijitter mechanism of HPA may cause an inadequate number of onetime scale-outs and, thus, the inability to satisfy subsequent service requests. The third concern is that the fixed data sampling time means that the time interval for data reporting is the same for average and high loads, leading to untimely and insufficient scaling at high load times. In this study, we propose a Double Threshold Horizontal Pod Autoscaler (DHPA) algorithm, which fine-grained divides the scale of events into three categories: scale-out, no scale, and scale-in. And then, on the scaling strength, we also employ two thresholds that are further subdivided into no scaling (antijitter), regular scaling, and fast scaling for each of the three cases. The DHPA algorithm determines the scaling strategy using the average of the growth rates of CPU utilization, and thus, different scheduling policies are adopted. We compare the DHPA with the HPA algorithm under different loads, including low, medium, and high. The experiments show that the DHPA algorithm has better antijitter and antiload characteristics in container increase and reduction while ensuring service and cluster security.

show abstract

Hierarchical Scaling of Microservices in Kubernetes

Cited by 41 publications

References 19 publications

Achieving AI-enabled Robust End-to-End Quality of Experience over Radio Access Networks

Achieving AI-enabled Robust End-to-End Quality of Experience over Radio Access Networks

Online Slice Reconfiguration for End-to-End QoE in 6G Applications

A Fine-Grained Horizontal Scaling Method for Container-Based Cloud

Contact Info

Product

Resources

About