Microservice-based Applications Research Articles

Optimized provisioning technique of future services with different QoS requirements in multi-access edge computing

With the present trend towards the use of microservices instead of monolithic software, future applications with diverse computational and delay-sensitive requirements, such as virtual reality (VR) or holographic communications (HC), will be redesigned based on loosely coupled microservices due to their more efficient deployment according to available resources. However, it remains a big challenge to provide such services under stringent quality of service (QoS) requirements at minimum time and cost and with an efficient utilization of the network. Presently, multi-access edge computing (MEC) has been proposed as a promising alternative to deploy future microservice-based applications to achieve faster response times while reducing the overall network load. However, network conditions, edge nodes state and service requests are dynamic, increasing the difficulty for the provisioning of services by MECs. In this context, we propose an optimization deployment heuristic (ODA) to obtain sub-optimal deployments in reasonable time, meeting the requirements of some representative future applications under study. In particular, according to latency requirements, there will be delay-tolerant and delay-sensitive applications, playing a major role regarding the alternatives for the deployment onto single or multiple MEC systems. The evaluation analysis shows that our proposal performs significantly better than other benchmark solutions in most of the metrics used for the experiments, especially in minimizing deployment costs in minimal time.

An artificial intelligence strategy for the deployment of future microservice-based applications in 6G networks

An artificial intelligence strategy for the deployment of future microservice-based applications in 6G networks

Root Cause Analysis in Microservice Using Neural Granger Causal Discovery

In recent years, microservices have gained widespread adoption in IT operations due to their scalability, maintenance, and flexibility. However, it becomes challenging for site reliability engineers (SREs) to pinpoint the root cause due to the complex relationship in microservices when facing system malfunctions. Previous research employed structure learning methods (e.g., PC-algorithm) to establish causal relationships and derive root causes from causal graphs. Nevertheless, they ignored the temporal order of time series data and failed to leverage the rich information inherent in the temporal relationships. For instance, in cases where there is a sudden spike in CPU utilization, it can lead to an increase in latency for other microservices. However, in this scenario, the anomaly in CPU utilization occurs before the latency increases, rather than simultaneously. As a result, the PC-algorithm fails to capture such characteristics. To address these challenges, we propose RUN, a novel approach for root cause analysis using neural Granger causal discovery with contrastive learning. RUN enhances the backbone encoder by integrating contextual information from time series and leverages a time series forecasting model to conduct neural Granger causal discovery. In addition, RUN incorporates Pagerank with a personalization vector to efficiently recommend the top-k root causes. Extensive experiments conducted on the synthetic and real-world microservice-based datasets demonstrate that RUN noticeably outperforms the state-of-the-art root cause analysis methods. Moreover, we provide an analysis scenario for the sock-shop case to showcase the practicality and efficacy of RUN in microservice-based applications. Our code is publicly available at https://github.com/zmlin1998/RUN.

A Federated Distributed Digital Forensic Readiness Model for the Cloud

Digital forensics in modern, cloud-based, microservice-based applications are complicated by multiple layers of abstraction, thereby making it difficult to accurately capture and correlate events that occur across these layers due to filtering caused by abstraction. The complexities linked to each layer of abstraction are primarily invisible to subsequent layers. Similarly, software services are often composed of one or more services provided by various service providers across the globe. Investigators are often faced with situations where breaches span over multiple service provider boundaries where not all digital forensic readiness evidence artefacts are captured by the service provider's forensic readiness processes. Instead, digital evidence artefacts are scattered across multiple service provider domains. This paper presents a novel, federated distributed digital forensic readiness model suitable for use in software-as-service, platform-as-service and infrastructure-as-service provider scenarios. The proposed model enables a service provider to capture and inspect forensic readiness artefacts in environments with various layers of abstraction. More importantly, the model also offers a way to share and access forensic readiness artefacts in a forensically sound manner to ultimately ensure that investigators can obtain a clear view of digital forensic events as they occur between amalgamated services provided by one or more separate service providers.

Cognitive complexity points: a metric to evaluate the design of microservices-based applications

The complexity of the software allows us to analyze how difficult to understand, implement and maintain the program can be. The metrics allow us to measure and estimate certain characteristics of the software to make decisions and corrective or preventive actions. The definition of the complexity of the microservices-based applications design is fundamental since it directly affects the performance of the application, development, testing, maintainability, storage (transactions and distributed queries), and the use and consumption of computational resources. In this paper, a cognitive complexity metric is proposed to evaluate the design and granularity of microservices-based applications, which define the required effort, or degree of difficulty to understand the microservices that make up the system. Typical cases were analyzed, which can appear in the design of microservices-based applications, the calculation of cognitive complexity was correct and consistent with the difficulty of understanding, maintaining, and developing a microservice system, therefore it is a viable option for analyzing complexity in microservices-based architecture.

PEMA+: A Comprehensive Resource Manager for Microservices

Motivation. Microservices architecture has become more prevalent in cloud-based applications where small, looselycoupled service components work together to handle user requests [1]. This architecture differs greatly from traditional monolithic deployments and provides advantages such as agile resource management, better fault tolerance, and platform compatibility [2, 3]. The figure 1 shows the difference between monolithic and microservice architecture. However, microservices come with their own set of challenges, particularly in the area of resource management. Resource management for microservices-based applications is more challenging than for monolithic applications due to the larger configuration space and complex communication topology [4, 5]. Additionally, existing resource management techniques for monolithic applications must manage microservices effectively. The resource management challenges for microservices must be addressed as more production cloud services adopt this architecture [6, 7].

Hybrelastic: A Hybrid Elasticity Strategy with Dynamic Thresholds for Microservice-based Cloud Applications

Hybrelastic: A Hybrid Elasticity Strategy with Dynamic Thresholds for Microservice-based Cloud Applications

Resource-aware Cyber Deception for Microservice-based Applications

Resource-aware Cyber Deception for Microservice-based Applications

Microservices performance forecast using dynamic Multiple Predictor Systems

Time series forecasting has been applied to predict performance degradation in Microservice-Based Applications (MBAs). The prediction enables MBA adaptation to avoid performance degradation and maintain the customer experience. The approaches in the literature commonly perform the forecast using a single monolithic model. However, since no model is better than others for all possible scenarios, using only one increases the risk of inaccurate estimates. Thus, an alternative for improving the accuracy and robustness of the performance degradation of MBAs is to adopt an ensemble, i.e., a Multiple Predictor Systems (MPS). This paper proposes an MPS methodology that selects and combines the most suitable models to forecast test patterns from a pool of models. Experiments were carried out with 32 time series containing performance metrics commonly used for MBA adaptation. Different scenarios concerning the pool (homogeneous or heterogeneous) and the selection phase (static or dynamic) are considered. Likewise, six widely used models in the literature were employed for pool generation: ARIMA, Multilayer Perceptron, Support Vector Regression, Random Forest, Long Short-Term Memory, and eXtreme Gradient Boosting. The evaluation shows that the proposed solution maintains or improves the forecast accuracy on 26 out of 32 datasets (81.25%) compared to monolithic models. In particular, the homogeneous pool and the dynamic selection of predictors obtained very satisfactory results. Therefore, MPS provide more accurate decision-making for MBA interventions, avoiding incorrect adjustments that could degrade the customer experience. Source code, figures, and data sets are publicly available at https://github.com/gfads/mps-methodology.

Diktyo: Network-Aware Scheduling in Container-Based Clouds

Containers have revolutionized application deployment and life-cycle management in current cloud platforms. Applications have evolved from single monoliths to complex graphs of loosely-coupled microservices. However, the efficient allocation of microservice-based applications is challenging due to their complex inter-dependencies. Further, recent applications are becoming even more delay-sensitive, demanding lower latency between dependent microservices. Scheduling policies in popular container orchestration platforms mainly aim to increase the resource efficiency of the infrastructure, insufficient for latency-sensitive applications. Application domains such as the Internet of Things and multi-tier web services would benefit from network-aware policies that consider network latency and bandwidth in the scheduling process. Previous works have studied network-aware scheduling via theoretical formulations or heuristic-based methods evaluated via simulations or small testbeds, making their full applicability in popular platforms difficult. This paper proposes a novel network-aware framework for the popular Kubernetes (K8s) platform named Diktyo that determines the placement of dependent microservices in long-running applications focused on reducing the application’s end-to-end latency and guaranteeing bandwidth reservations. Simulations show that Diktyo can significantly reduce the network latency for various applications across different infrastructure topologies compared to default K8s scheduling plugins. Also, experiments in a K8s cluster with microservice benchmark applications show that Diktyo can increase database throughput by 22% and reduce application response time by 45%.

User Authorization in Microservice-Based Applications

Microservices have emerged as a prevalent architectural style in modern software development, replacing traditional monolithic architectures. The decomposition of business functionality into distributed microservices offers numerous benefits, but introduces increased complexity to the overall application. Consequently, the complexity of authorization in microservice-based applications necessitates a comprehensive approach that integrates authorization as an inherent component from the beginning. This paper presents a systematic approach for achieving fine-grained user authorization using Attribute-Based Access Control (ABAC). The proposed approach emphasizes structure preservation, facilitating traceability throughout the various phases of application development. As a result, authorization artifacts can be traced seamlessly from the initial analysis phase to the subsequent implementation phase. One significant contribution is the development of a language to formulate natural language authorization requirements and policies. These natural language authorization policies can subsequently be implemented using the policy language Rego. By leveraging the analysis of software artifacts, the proposed approach enables the creation of comprehensive and tailored authorization policies.

Cost-Efficient Fault-Tolerant Workflow Scheduling for Deadline-Constrained Microservice-Based Applications in Clouds

Microservices are becoming increasingly popular in the construction of cloud applications. On the basis of containers, microservice instances can be implemented with high scalability and maintainability. Due to the need of ensuring various quality of service (QoS) requirements and the two-layer resource structure of containers and virtual machines (VMs), microservice workflow scheduling in clouds is a challenging problem to address. This paper proposes a heuristic algorithm GSMS to minimize execution cost of a microservice-based workflow application while satisfying deadline and reliability constraints. GSMS adopts a greedy fault-tolerant scheduling strategy for replicas of each task to select appropriate resources that meet the sub-deadline and minimize the cost until the sub-reliability is guaranteed. Furthermore, a resource adjustment strategy is incorporated into GSMS to further improve resource utilization. By conducting extensive experiments with several realistic workflow applications, in comparison with existing algorithms, the effectiveness and efficiency of GSMS in achieving lower execution cost and meeting deadline and reliability requirements are validated.

Performance Analysis of Microservices Behavior in Cloud vs Containerized Domain based on CPU Utilization

Enterprise application development is rapidly moving towards a microservices-based approach. Microservices development makes application deployment more reliable and responsive based on their architecture and the way of deployment. Still, the performance of microservices is different in all environments based on resources provided by the respective cloud and services provided in the backend such as auto-scaling, load balancer, and multiple monitoring parameters. So, it is strenuous to identify Scaling and monitoring of microservice-based applications are quick as compared to monolithic applications [1]. In this paper, we deployed microservice applications in cloud and containerized environments to analyze their CPU utilization over multiple network input requests. Monolithic applications are tightly coupled while microservices applications are loosely coupled which help the API gateway to easily interact with each service module. With reference to monitoring parameters, CPU utilization is 23 percent in cloud environment. Additionally, we deployed the equivalent microservice in a containerized environment with extended resources to minimize CPU utilization to 17 percent. Furthermore, we have shown the performance of the application with “Network IN” and “Network Out” requests.

Advanced data analytics using three-stage intelligent model pipelining for containerized microservices in 5G networks and beyond

The recent rapid advancement of cloud-native networking infrastructure has leveraged the resource virtualization technology of containers to realize diverse microservice-based applications in 5G/6G networks and clouds. Containers drastically enhance the efficiency of computational resource allocation and utilization as compared to the related virtualization technology of Virtual Machines (VMs). The networking environment leveraging both VM and container mixed virtualization technologies makes the most use of them to realize a computational platform whose resources can be dynamically adjusted to a fine granularity. To continuously meet the required levels of quality of services in 5G/6G networks and clouds in that platform, an agile and autonomous data analytics system in the control plane is essential for the accurate prediction of server workloads and dynamic allocation of enough amount of computational resource. In this paper, we introduce a framework, which complies with Recommendation ITU-T Y.3177, for autonomous computational resource control and management. The framework consists of an advanced data analytics system and a resource control system. We propose an architecture for the advanced data analytics system consisting of learning and prediction components. The learning component includes a three-stage intelligent model pipelining with three cascaded machine learning models, nonlinear regression, clustering, and multiple regression. These models determine the fluctuation trends in CPU utilization, classify services with similarities in the trends, and predict the peak CPU utilization of each containerized microservice. We evaluate the proposed models through experiments and numerical analysis. The results prove that the models support agile data analytics, which can complete data processing in the time granularity of seconds and achieve higher prediction accuracy of CPU utilization.

SEMGROMI-a semantic grouping algorithm to identifying microservices using semantic similarity of user stories.

Microservices is an architectural style for service-oriented distributed computing, and is being widely adopted in several domains, including autonomous vehicles, sensor networks, IoT systems, energy systems, telecommunications networks and telemedicine systems. When migrating a monolithic system to a microservices architecture, one of the key design problems is the "microservice granularity definition", i.e., deciding how many microservices are needed and allocating computations among them. This article describes a semantic grouping algorithm (SEMGROMI), a technique that takes user stories, a well-known functional requirements specification technique, and identifies number and scope of candidate microservices using semantic similarity of the user stories' textual description, while optimizing for low coupling, high cohesion, and high semantic similarity. Using the technique in four validation projects (two state-of-the-art projects and two industry projects), the proposed technique was compared with domain-driven design (DDD), the most frequent method used to identify microservices, and with a genetic algorithm previously proposed as part of the Microservices Backlog model. We found that SEMGROMI yields decompositions of user stories to microservices with high cohesion (from the semantic point of view) and low coupling, the complexity was reduced, also the communication between microservices and the estimated development time was decreased. Therefore, SEMGROMI is a viable option for the design and evaluation of microservices-based applications. The proposed semantic similarity-based technique (SEMGROMI) is part of the Microservices Backlog model, which allows to evaluate candidate microservices graphically and based on metrics to make design-time decisions about the architecture of the microservices-based application.

LWS: A framework for log-based workload simulation in session-based SUT

Artificial intelligence for IT Operations (AIOps) plays a critical role in operating and managing cloud-native systems and microservice-based applications but is limited by the lack of high-quality datasets with diverse scenarios. Realistic workloads are the premise and basis of generating such AIOps datasets, with the session-based workload being one of the most typical examples. Due to privacy concerns, complexity, variety, and requirements for reasonable intervention, it is difficult to copy or generate such workloads directly, showing the importance of effective and intervenable workload simulation. In this paper, we formulate the task of workload simulation and propose a framework for Log-based Workload Simulation (LWS) in session-based systems. LWS extracts the workload specification including the user behavior abstraction based on agglomerative clustering as well as relational models and the intervenable workload intensity from session logs. Then LWS combines the user behavior abstraction with the workload intensity to generate simulated workloads. The experimental evaluation is performed on an open-source cloud-native application with both well-designed and public real-world workloads, showing that the simulated workload generated by LWS is effective and intervenable, which provides the foundation of generating high-quality AIOps datasets.

PerfSim: A Performance Simulator for Cloud Native Microservice Chains

Cloud native computing paradigm allows microservice-based applications to take advantage of cloud infrastructure in a scalable, reusable, and interoperable way. However, in a cloud native system, the vast number of configuration parameters and highly granular resource allocation policies can significantly impact the performance and deployment cost. For understanding and analyzing these implications in an easy, quick, and cost-effective way, we present PerfSim, a discrete-event simulator for approximating and predicting the performance of cloud native service chains in user-defined scenarios. To this end, we proposed a systematic approach for modeling the performance of microservices endpoint functions by collecting and analyzing their performance and network traces. With a combination of the extracted models and user-defined scenarios, PerfSim can then simulate the performance behavior of all services over a given period and provide an approximation for system KPIs, such as requests' average response time. Using the processing power of a single laptop, we evaluated both simulation accuracy and speed of PerfSim in 104 prevalent scenarios and compared the simulation results with the identical deployment in a real Kubernetes cluster. We achieved ~81-99% simulation accuracy in approximating the average response time of incoming requests and ~16-1200 times speed-up factor for the simulation.

Kubernetes Scheduling: Taxonomy, Ongoing Issues and Challenges

Continuous integration enables the development of microservices-based applications using container virtualization technology. Container orchestration systems such as Kubernetes, which has become the de facto standard, simplify the deployment of container-based applications. However, developing efficient and well-defined orchestration systems is a challenge. This article focuses specifically on the scheduler, a key orchestrator task that assigns physical resources to containers. Scheduling approaches are designed based on different Quality of Service (QoS) parameters to provide limited response time, efficient energy consumption, better resource utilization, and other things. This article aims to establish insight knowledge into Kubernetes scheduling, find the main gaps, and thus guide future research in the area. Therefore, we conduct a study of empirical research on Kubernetes scheduling techniques and present a new taxonomy for Kubernetes scheduling. The challenges, future direction, and research opportunities are also discussed.

A Survey on Graph Neural Networks for Microservice-Based Cloud Applications

Graph neural networks (GNNs) have achieved great success in many research areas ranging from traffic to computer vision. With increased interest in cloud-native applications, GNNs are increasingly being investigated to address various challenges in microservice architecture from prototype design to large-scale service deployment. To appreciate the big picture of this emerging trend, we provide a comprehensive review of recent studies leveraging GNNs for microservice-based applications. To begin, we identify the key areas in which GNNs are applied, and then we review in detail how GNNs can be designed to address the challenges in specific areas found in the literature. Finally, we outline potential research directions where GNN-based solutions can be further applied. Our research shows the popularity of leveraging convolutional graph neural networks (ConGNNs) for microservice-based applications in the current design of cloud systems and the emerging area of adopting spatio-temporal graph neural networks (STGNNs) and dynamic graph neural networks (DGNNs) for more advanced studies.

Microservices-Adaptive Software-Defined Load Balancing for 5G and Beyond Ecosystems

Although 5G networks achieve impressive improvements in terms of latency, data rates and network capacity, their evolution requires the synergy of networks, clouds and applications. For example, targeted challenging applications require efficient end-to-end performance, impacted by emerging technologies beyond radio, including: Software-Defined Networks (SDNs) enabling programmability in the network environment from a centralized viewpoint; cloud orchestrators dynamically adapting server resource allocation to dynamic workloads; and microservice-based applications consisting of minimal, single-purpose service functions communicating with each other, individually implemented and scaled towards efficient resource allocation and fault-tolerance. In this context, we propose an SDN load balancing solution for a specific class of services consisting of microservices with heterogeneous but simpler network and server resource requirements, compared to the resource-demands of the whole service. Our proposal adapts to the estimated resource demands of individual microservices, balances resource utilization among cloud servers and improves service operation in terms of response times, throughput and fairness. We validate experimentally the proposed platform and its mechanisms with a particular 5G and beyond use-case with the above characteristics.