Jul 2026· Fall Joint Computer Conference· pp. 438-443· 0 citations· 17 references
Abstract
Modern cloud-native microservice systems face significant challenges when deployed in ultra-large-scale, missioncritical environments such as civil aviation passenger service platforms. Existing SDK-based frameworks suffer from tight coupling and limited cross-language flexibility, while Service Mesh solutions introduce scalability bottlenecks due to globally synchronized service state and configuration overhead. This paper presents a partition-based microservice architecture that introduces a hierarchical control model separating local traffic interception from global routing intelligence. The system is built upon a decoupled Sidecar-Router design, where Sidecar Proxies handle lightweight request interception and partition-level forwarding, while Service Routers execute instance-level routing and traffic governance. By introducing partition-bounded service discovery and physical-IP-based cross-cluster communication, the proposed architecture eliminates global state synchronization of high-frequency service metadata and ensures constant resource overhead at the data plane. The system has been deployed in a large civil aviation PSS environment, supporting over 1,800 microservices and more than 6,000 instances, with a sustained production throughput of over 100,000 TPS and 60% business coverage via the traffic access platform. The results demonstrate that partitioned routing and decoupled control significantly improve scalability, fault isolation, and operational efficiency in large-scale distributed microservice systems.
The paper addresses the transformation of enterprise application infrastructure out of on-premise legacy resource setting into service-based cloud environments properly configured to scale horizontally, and presents experimental evaluations of the response time, throughput, service resiliency, and infrastructure utilization in both traditional and cloud-native deployments.
Kanya Mohammed, Naree Thongchai· International Journal of Mod...· 0 citations
The proposed approach separates the research-infrastructure layer, which exposes and manages distributed resources, from the application layer, where Cyber-Physical workflows are organized according to an Edge-Fog-Cloud pattern in which placement, timing, and data provenance are treated as first-class experimental concerns.
Fabio Orazio Mirto, Giuseppe Tricomi, L. D’Agati et al.· arXiv.org· 0 citations
As computing resources in cloud environments become increasingly abundant, executing complex scientific workflows on large-scale cloud infrastructure has become a standard practice. However, communication-intensive workflows face two fundamental bottlenecks. First, the lack of physical topology awareness often forces high-frequency interacting microservices to be placed on geographically distant nodes, which generates excessive cross-node communication overhead, leads to network load imbalance, and increases latency. Second, the prohibitive online computation time of conventional iterative scheduling algorithms further degrades response speed, making them unsuitable for real-time scenarios. To address these bottlenecks, this paper proposes AMSche, a framework for microservice deployment and task scheduling that is aware of both position and topology. The framework comprises two core mechanisms. The first mechanism, position-aware service deployment, colocates high-frequency interacting services on the same physical node based on communication affinity, thereby compressing cross-node communication overhead at the physical level. The second mechanism, topology-aware task scheduling, leverages online topology feature similarity mapping to instantly reuse historical scheduling plans, achieving scheduling decisions at the millisecond level. Extensive experiments on real-world scientific workflow datasets demonstrate that AMSche achieves an average improvement of 16.19% to 39.15% over existing baseline methods in comprehensive metrics including response time, total communication volume, and network load balance.
Hao Wei, Hailiang Chen, Jia-Nan Sun et al.· Fall Joint Computer Conferen...· 0 citations
Aim: This study aimed to design and evaluate a cloud-native reference architecture for real-time payment processing that integrates microservices, event-driven communication, infrastructure-as-code, and federated AIOps to improve scalability, resilience, regulatory compliance, and operational efficiency. The novelty of the study is the federated three-tier integration model that combines ITIL-aligned governance, DevOps continuous delivery, and AIOps-driven anomaly detection within a single payment-specific reference architecture.
Methods: A mixed-methods design was employed, combining a structured synthesis of peer-reviewed and industry literature with quantitative benchmarking of instrumented cloud-native reference deployments. The deployments were scaled from 1 to 128 service replicas across two public clouds and an on-premise environment and were evaluated against a representative monolithic baseline.
Results: The findings show that the proposed cloud-native architecture reduced end-to-end p99 latency from approximately 850 ms in the monolithic baseline to approximately 180 ms, representing a 4.7× improvement. Sustained throughput scaled near-linearly to 78,000 transactions per second across 128 service replicas, compared with approximately 3,800 transactions per second in the baseline. Federated AIOps reduced Mean Time to Detect (MTTD) by 60% and Mean Time to Resolve (MTTR) by 50%, while alert volume and false-positive rates decreased by 70% and 82%, respectively. The architecture achieved 99.99% availability, with a Recovery Time Objective (RTO) of less than five minutes and a Recovery Point Objective (RPO) of less than 30 seconds across multi-region active-active deployments. The study also identifies operational complexity, multi-cloud cost overhead, and managed-service vendor lock-in as key limitations.
Conclusion: The findings indicate that cloud-native architectures can substantially improve the scalability, operational resilience, and regulatory readiness of real-time payment systems compared with conventional monolithic designs.
Recommendation: Financial institutions planning payment modernization should consider a phased adoption of cloud-native architectures supported by integrated governance, automation, and observability practices. The proposed framework should also be validated across diverse production payment environments.
Midhun Michael Nelavala· American Journal of Technolo...· 0 citations
This work presents a comprehensive overview of the TSN deployment lifecycle, current challenges, limitations of existing tools, and future research directions for TSN deployment and management, and identifies key research gaps from a deployment perspective and provides guidance for the development of next-generation deployable TSN networks.
Rubi Debnath, Paul Pop, Silviu S. Craciunas et al.· 0 citations
APIs Become the Primary Communication, Orchestration and Business Operations Channel in Cloud-Native Microservices. While this architecture offers better scalability, flexibility and agility of deployment, it also increases the security risk by introducing the distributed trust boundary issue (east-west traffic), short-lived workloads and highly-permeable service-to-service communication. Traditional API security methods like perimeter filtering, signature-based detection and isolated identity checks have become insufficient for these highly dynamic environments. To overcome this challenge, and propose a Cognitive Security Framework for Cloud-Native Microservices APIs (CSF-API). The framework integrates zero-trust, workload identity, policy-as-code, service mesh attestation and enforcement, Kubernetes-native security controls and telemetry-driven intelligence into a singular model. It keeps an eye on API activity, analyzes request context through identity and runtime signals, and enforces appropriate responses (allow, audit, rate-limit, challenge, isolate or deny). The suggested architecture includes API gateways, external authorization with Envoy, Open Policy Agent and SPIFFE/SPIRE identities along with Istio authorization features, Kubernetes admission control/pod security features and observability by OpenTelemetry. Traditional reactive API protection is no longer viable due to stricter global cloud infrastructures and massive cloud-native microservice applications.
Makeshwari Sattanadin· International Conference Com...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
