Performance Analysis of Architectural Patterns for Federated Learning Systems
Fri 4 Apr 2025 10:30 - 10:45 at Hall 2 (U82) - Software Architecture Experimentation and Practice I Chair(s): Sebastian Werner
Designing Federated Learning systems is not trivial, as it requires managing heterogeneous and distributed clients’ resources, while balancing data privacy and system efficiency. Architectural patterns have been recently specified in the literature to showcase reusable solutions to common problems within Federated Learning systems. However, patterns often lead to both benefits and drawbacks, e.g., introducing a message compressor algorithm may reduce the system communication time, but it may produce additional computational costs for clients’ devices. The goal of this paper is to quantitatively investigate the performance impact of applying a selected set of architectural patterns when designing Federated Learning systems, thus providing evidence of their pros and cons. We develop an open source environment by extending the well-established Flower framework; it integrates the implementation of four architectural patterns and evaluates their performance characteristics. Experimental results assess that architectural patterns indeed bring performance gains and pains, as raised by the practitioners in the literature. Our framework can support software architects in making informed design choices when designing Federated Learning systems
Thu 3 AprDisplayed time zone: Brussels, Copenhagen, Madrid, Paris change
Fri 4 AprDisplayed time zone: Brussels, Copenhagen, Madrid, Paris change
10:30 - 11:30 | Software Architecture Experimentation and Practice INew and Emerging Ideas / Software Architecture in Practice / Research Papers at Hall 2 (U82) Chair(s): Sebastian Werner TU Berlin, Germany | ||
10:30 15mResearch paper | Performance Analysis of Architectural Patterns for Federated Learning Systems Research Papers Ivan Compagnucci Gran Sasso Science Institute, Riccardo Pinciroli Gran Sasso Science Institute, Catia Trubiani Gran Sasso Science Institute Pre-print | ||
10:45 15mPaper | An Architecture and Protocol for Decentralized Retrieval Augmented Generation New and Emerging Ideas Tobias Hecking , Thorsten Sommer , Michael Felderer German Aerospace Center (DLR) & University of Cologne | ||
11:00 15mPaper | Fast and Efficient What-If Analyses of Invocation Overhead and Transactional Boundaries to Support the Migration to Microservices New and Emerging Ideas Pre-print | ||
11:15 15mPaper | Extensible Experimentation Platform: Effective A/B Test Analysis at Scale Software Architecture in Practice |