[论文解读] A Microservice-Based Platform for Sustainable and Intelligent SLO Fulfilment and Service Management

Juan Luis Herrera, Daniel Wang+1 more|arXiv (Cornell University)|Feb 13, 2026

Software System Performance and Reliability被引用 0

一句话总结

论文提出了 CASCA，一种开源的基于微服务的平台，能够实现碳感知的 SLO 履行与动态服务管理，同时保护开发者隐私，并在真实 CC 媒体流媒体用例中进行演示，决策系统使用 Bash、Rust 和 Python 实现。

ABSTRACT

The Microservices Architecture (MSA) design pattern has become a staple for modern applications, allowing functionalities to be divided across fine-grained microservices, fostering reusability, distribution, and interoperability. As MSA-based applications are deployed to the Computing Continuum (CC), meeting their Service Level Objectives (SLOs) becomes a challenge. Trading off performance and sustainability SLOs is especially challenging. This challenge can be addressed with intelligent decision systems, able to reconfigure the services during runtime to meet the SLOs. However, developing these agents while adhering to the MSA pattern is complex, especially because CC providers, who have key know-how and information to fulfill these SLOs, must comply with the privacy requirements of application developers. This work presents the Carbon-Aware SLO and Control plAtform (CASCA), an open-source MSA-based platform that allows CC providers to reconfigure services and fulfill their SLOs while maintaining the privacy of developers. CASCA is architected to be highly reusable, distributable, and easy to use, extend, and modify. CASCA has been evaluated in a real CC testbed for a media streaming service, where decision systems implemented in Bash, Rust, and Python successfully reconfigured the service, unaffected by upholding privacy.

研究动机与目标

• Address the challenge of fulfilling SLOs in microservice-based applications deployed in the Computing Continuum while considering sustainability and privacy.
• Provide a reusable, extensible platform (CASCA) that enables carbon-aware SLO fulfilment and service management across distributed environments.
• Integrate carbon-intensity data via a dedicated EMMA microservice and offer privacy-preserving interfaces for infrastructure providers and developers.
• Demonstrate feasibility and practicality through a real CC testbed using a media streaming use case and open-source implementations.

提出的方法

• Proposes the CASCA platform architecture following microservice (MSA) principles to support SLO fulfilment and service management.
• Introduces EMMA, a carbon-intensity API to enable carbon-awareness as part of decision making.
• Defines a dual-API service API (SLO API and service control API) with privacy-preserving mappings between SLOs, configurations, and infrastructure observability.
• Provides a telemetry and observability layer comprising telemetry middleware, telemetry hooks, and a telemetry database to unify multi-source SLO data.
• Outlines integration with decision systems (AI-based or traditional) via OpenAPI-compliant interfaces and supports online training requirements.
• Details a media streaming use case implemented with Jellyfin, Autowatcher (Salesforce-like continuous reporter), MQTT-based telemetry, and InfluxDB as the time-series data store, all containerized for MSA deployment.

Figure 1 : Complete architecture of the CASCA platform

实验结果

研究问题

• RQ1 How can CASCA enable coexisting SLO fulfilment and service management in an MSAs deployed on the CC without leaking sensitive service information to infrastructure providers?
• RQ2 Can carbon-awareness be effectively integrated into SLO fulfilment through a dedicated EMMA microservice and standard APIs?
• RQ3 What is the practicality and effectiveness of deploying decision systems of different implementations (shell/Bash, Rust, Python) within CASCA to reconfigure services at runtime?
• RQ4 Does the CASCA architecture support privacy-preserving observability and flexible data source integration for diverse telemetry?
• RQ5 Is the CASCA deployment in a real CC testbed able to manage trade-offs between performance SLOs (e.g., transcoding throughput) and carbon footprint in a media streaming scenario?

主要发现

• CASCA enables reconfiguration and observability of SLO fulfilment in a privacy-preserving manner using modular microservices.
• An open API gateway with SLO and service control interfaces supports diverse decision-system implementations while decoupling infrastructure from service semantics.
• EMMA provides carbon-intensity data through OpenAPI-compliant endpoints enabling carbon-aware decisions.
• The media streaming use case demonstrates CASCA’s ability to reconfigure services (e.g., transcoding) while accounting for carbon footprint, with decision systems implemented in Bash, Rust, and Python.
• All components and use-case implementations are open source, promoting replication and reuse in other MSAs and CC deployments.

Figure 2 : Implementation of CASCA for the media streaming use case.