Resource Allocation With Workload-Time Windows for Cloud-Based Software Services: A Deep Reinforcement Learning Approach

Chen, Xing; Yang, Lijian; Chen, Zheyi; Min, Geyong; Zheng, Xianghan; Rong, Chunming

doi:10.1109/tcc.2022.3169157

Cited by 24 publications

(18 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this review, the factor is referring to the aspect addressed in the existing works while the metric refers to the metric used to make the scaling decision. Most of the studies work on the system performance by specifically addressing the performance indicator such as response time, 78 throughput, 30 availability, 39 accuracy, 34 workloads, 82 latency, 74,76 and energy 83 . The next factor that is mostly concerned the researcher is the cost of renting the machine or resources.…”

Section: Comparison Studymentioning

confidence: 99%

“…The verified results can help the autoscaling process to decide the best possible scaling action which follows the requirements of the microservices. Instead of determining the correctness of the selected scaling action, the existing works utilize the ML techniques for different purposes such as to enhance the process of identifying the threshold values of autoscaling, 32 predict the load of microservices, 36 predict management operation 82 …”

Section: Comparison Studymentioning

confidence: 99%

“…However, the implementation might be tricky if the scaling approach involves different types of programming language. The recent implementation of DRL‐based cloud resource allocation is performed by Chen et al 82 while the works that implement verification for DRL‐based autoscaling are barely found. Verification of transfer learning‐based autoscaling : The transfer learning (TL) is an approach that involves the reuse of a pretrained model to train another model to solve different task 97 . Meanwhile, TL is not a new thing for cloud computing but the existing works that include the implementation of formal verification is hardly found.…”

Section: Open Challenges and Opportunitiesmentioning

confidence: 99%

“…76 Apart from these techniques, there are several combinations of approaches used to enhance the scaling algorithm by optimizing the algorithm, 31,77,79 predicting the burstiness of workloads, 72 implementing two-layer scaling, 73 implementing Green-aware scheduling to scale the hosts 83 and implementing an agnostic controller module into the scaling algorithm. 27 Meanwhile, the work done by Chen et al 82 implements a DRL algorithm whilst another work had combined one of the common scaling method (i.e., Queuing Theory) with formal method. 7 The next observed elements are the factors and metrics which closely related to each other.…”

Section: Ta B L Ementioning

confidence: 99%

“…Instead of determining the correctness of the selected scaling action, the existing works utilize the ML techniques for different purposes such as to enhance the process of identifying the threshold values of autoscaling, 32 predict the load of microservices, 36 predict management operation. 82…”

Section: Ta B L Ementioning

confidence: 99%

See 4 more Smart Citations

A review of microservices autoscaling with formal verification perspective

2022

View full text Add to dashboard Cite

The process of scaling microservices is a challenging task, especially in maintaining optimum resource provisioning while respecting QoS constraints and SLA.Many research works have proposed autoscaling approaches for microservices, however, less likely concerned with the correctness guarantee of the proposed algorithms. Hence, it is significant to gather and summarize these approaches to foster future innovation. Meanwhile, a few reviews have been published concerning microservices from different aspects. Therefore, our review complements the existing by focusing on autoscaling with verification perspectives. This study highlights the recent contributions in three inter-related main topics that were published within the year 2017 to 2022, namely, microservice, verification, and autoscaling. Due to limited resources on verification for microservice autoscaling, we widen the perspective by considering the verification for autoscaling in cloud-based systems. Based on our findings, we found that the formal method is not a new thing in verifying the autoscaling policies in cloud-based systems, and one recent study that implements the formal method in the microservices area has been identified. Apart from the autoscaling techniques, we have also determined several factors that have been a concern in scaling the microservices as well as the relatable metrics. Meanwhile, from a verification perspective, we identified that probabilistic model checking is the common formal verification technique used to verify microservices and cloud autoscaling. Finally, we recommend open challenges from two perspectives which highlight the verification for existing microservice autoscaling and verification for ML-based microservice autoscaling.

show abstract

Section: Comparison Studymentioning

confidence: 99%

Section: Comparison Studymentioning

confidence: 99%

Section: Open Challenges and Opportunitiesmentioning

confidence: 99%

Section: Ta B L Ementioning

confidence: 99%

Section: Ta B L Ementioning

confidence: 99%

See 3 more Smart Citations

A review of microservices autoscaling with formal verification perspective

2022

View full text Add to dashboard Cite

show abstract

Evaluation of soft computing in methodology for calculating information protection from parameters of its distribution in social networks

et al. 2023

View full text Add to dashboard Cite

Deep reinforcement learning-based methods for resource scheduling in cloud computing: a review and future directions

Zhou,

Tian,

Buyya

et al. 2024

Artif Intell Rev

View full text Add to dashboard Cite

With the acceleration of the Internet in Web 2.0, Cloud computing is a new paradigm to offer dynamic, reliable and elastic computing services. Efficient scheduling of resources or optimal allocation of requests is one of the prominent issues in emerging Cloud computing. Considering the growing complexity of Cloud computing, future Cloud systems will require more effective resource management methods. In some complex scenarios with difficulties in directly evaluating the performance of scheduling solutions, classic algorithms (such as heuristics and meta-heuristics) will fail to obtain an effective scheme. Deep reinforcement learning (DRL) is a novel method to solve scheduling problems. Due to the combination of deep learning and reinforcement learning (RL), DRL has achieved considerable performance in current studies. To focus on this direction and analyze the application prospect of DRL in Cloud scheduling, we provide a comprehensive review for DRL-based methods in resource scheduling of Cloud computing. Through the theoretical formulation of scheduling and analysis of RL frameworks, we discuss the advantages of DRL-based methods in Cloud scheduling. We also highlight different challenges and discuss the future directions existing in the DRL-based Cloud scheduling.

show abstract

Resource Allocation With Workload-Time Windows for Cloud-Based Software Services: A Deep Reinforcement Learning Approach

Cited by 24 publications

References 34 publications

A review of microservices autoscaling with formal verification perspective

A review of microservices autoscaling with formal verification perspective

Evaluation of soft computing in methodology for calculating information protection from parameters of its distribution in social networks

Deep reinforcement learning-based methods for resource scheduling in cloud computing: a review and future directions

Contact Info

Product

Resources

About