Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Microservice-Based Platform for Sustainable and Intelligent SLO Fulfilment and Service Management

Juan Luis Herrera, Daniel Wang, Schahram Dustdar

TL;DR

CASCA introduces a carbon-aware, open-source platform for SLO fulfilment and service management in microservice architectures deployed across the Computing Continuum, addressing privacy between service developers and infrastructure providers. It combines an EMMA-based carbon-intensity data layer with a modular SLO/telemetry framework and a flexible, API-driven Service API to support diverse decision systems, including AI-based approaches. The authors validate CASCA with a media streaming use case, implementing RDS, GDS, and RLDS decision systems that reconfigure transcoding parameters to balance FPS and carbon footprint, demonstrating privacy preservation and rapid reconfiguration. The results indicate CSCA enables effective SLO management with modest overhead, supporting both traditional and AI-driven decision systems, and point to future work in hardware control, CC-wide orchestration, and developer-friendly dashboards.

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.

A Microservice-Based Platform for Sustainable and Intelligent SLO Fulfilment and Service Management

TL;DR

CASCA introduces a carbon-aware, open-source platform for SLO fulfilment and service management in microservice architectures deployed across the Computing Continuum, addressing privacy between service developers and infrastructure providers. It combines an EMMA-based carbon-intensity data layer with a modular SLO/telemetry framework and a flexible, API-driven Service API to support diverse decision systems, including AI-based approaches. The authors validate CASCA with a media streaming use case, implementing RDS, GDS, and RLDS decision systems that reconfigure transcoding parameters to balance FPS and carbon footprint, demonstrating privacy preservation and rapid reconfiguration. The results indicate CSCA enables effective SLO management with modest overhead, supporting both traditional and AI-driven decision systems, and point to future work in hardware control, CC-wide orchestration, and developer-friendly dashboards.

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.
Paper Structure (17 sections, 3 equations, 11 figures, 1 table, 1 algorithm)

This paper contains 17 sections, 3 equations, 11 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related works
  3. CASCA architecture
  4. SLOs and telemetry
  5. Service API
  6. Decision system integration
  7. Use case: Media streaming service
  8. Use case description
  9. Implementation details
  10. Decision systems
  11. Random Decision System
  12. Greedy Decision System
  13. Reinforcement Learning Decision System
  14. Evaluation
  15. Evaluation setup
  16. ...and 2 more sections

Figures (11)

  • Figure 1: Complete architecture of the CASCA platform
  • Figure 2: Implementation of CASCA for the media streaming use case.
  • Figure 3: Average frames per second achieved over time with all decision systems and full descriptions.
  • Figure 4: Power consumption over time with all decision systems and full descriptions.
  • Figure 5: Carbon footprint over time with all decision systems and full descriptions.
  • ...and 6 more figures