arXiv subjects

Chushu Gao

Publications and source records attributed to Chushu Gao.

3 recordsLinked to original sources

arXiv2026

On the Effectiveness of Proposed Techniques to Reduce Energy Consumption in RAG Systems: A Controlled Experiment

The rising energy demands of machine learning (ML), e.g., implemented in popular variants like retrieval-augmented generation (RAG) systems, have raised significant concerns about their environmental sustainability. While previous research has proposed green tactics for ML-enabled systems, their empirical evaluation within RAG systems remains largely unexplored. This study presents a controlled experiment investigating five practical techniques aimed at reducing energy consumption in RAG systems. Using a production-like RAG system developed at our collaboration partner, the Software Improvement Group, we evaluated the impact of these techniques on energy consumption, latency, and accuracy. Through a total of 9 configurations spanning over 200 hours of trials using the CRAG dataset, we reveal that techniques such as increasing similarity retrieval thresholds, reducing embedding sizes, applying vector indexing, and using a BM25S reranker can significantly reduce energy usage, up to 60% in some cases. However, several techniques also led to unacceptable accuracy decreases, e.g., by up to 30% for the indexing strategies. Notably, finding an optimal retrieval threshold and reducing embedding size substantially reduced energy consumption and latency with no loss in accuracy, making these two techniques truly energy-efficient. We present the first comprehensive, empirical study on energy-efficient design techniques for RAG systems, providing guidance for developers and researchers aiming to build sustainable RAG applications.

cs.SE↗

arXiv2025

On the Effectiveness of Microservices Tactics and Patterns to Reduce Energy Consumption: An Experimental Study on Trade-Offs

Context: Microservice-based systems have established themselves in the software industry. However, sustainability-related legislation and the growing costs of energy-hungry software increase the importance of energy efficiency for these systems. While some proposals for architectural tactics and patterns exist, their effectiveness as well as potential trade-offs on other quality attributes (QAs) remain unclear. Goal: We therefore aim to study the effectiveness of microservices tactics and patterns to reduce energy consumption, as well as potential trade-offs with performance and maintainability. Method: Using the open-source Online Boutique system, we conducted a controlled experiment with three tactics and three patterns, and analyzed the impact of each technique compared to a baseline. We also tested with three levels of simulated request loads (low, medium, high). Results: Request load moderated the effectiveness of reducing energy consumption. All techniques (tactics and patterns) reduced the energy consumption for at least one load level, up to 5.6%. For performance, the techniques could negatively impact response time by increasing it by up to 25.9%, while some also decreased it by up to 72.5%. Two techniques increased the throughput, by 1.9% and 34.0%. For maintainability, three techniques had a negative, one a positive, and two no impact. Conclusion: Some techniques reduced energy consumption while also improving performance. However, these techniques usually involved a trade-off in maintainability, e.g., via more code duplication and module coupling. Overall, all techniques significantly reduced energy consumption at higher loads, but most of them sacrificed one of the other QAs. This highlights that the real challenge is not simply reducing energy consumption of microservices, but to achieve energy efficiency.

cs.SE↗

arXiv2017

SpreadCluster: Recovering Versioned Spreadsheets through Similarity-Based Clustering

Version information plays an important role in spreadsheet understanding, maintaining and quality improving. However, end users rarely use version control tools to document spreadsheet version information. Thus, the spreadsheet version information is missing, and different versions of a spreadsheet coexist as individual and similar spreadsheets. Existing approaches try to recover spreadsheet version information through clustering these similar spreadsheets based on spreadsheet filenames or related email conversation. However, the applicability and accuracy of existing clustering approaches are limited due to the necessary information (e.g., filenames and email conversation) is usually missing. We inspected the versioned spreadsheets in VEnron, which is extracted from the Enron Corporation. In VEnron, the different versions of a spreadsheet are clustered into an evolution group. We observed that the versioned spreadsheets in each evolution group exhibit certain common features (e.g., similar table headers and worksheet names). Based on this observation, we proposed an automatic clustering algorithm, SpreadCluster. SpreadCluster learns the criteria of features from the versioned spreadsheets in VEnron, and then automatically clusters spreadsheets with the similar features into the same evolution group. We applied SpreadCluster on all spreadsheets in the Enron corpus. The evaluation result shows that SpreadCluster could cluster spreadsheets with higher precision and recall rate than the filename-based approach used by VEnron. Based on the clustering result by SpreadCluster, we further created a new versioned spreadsheet corpus VEnron2, which is much bigger than VEnron. We also applied SpreadCluster on the other two spreadsheet corpora FUSE and EUSES. The results show that SpreadCluster can cluster the versioned spreadsheets in these two corpora with high precision.

cs.SE↗