Memory failures in large-scale production environments pose critical threats to system reliability and service availability. While existing studies have conducted in-depth analyses of the temporal and spatial correlations of memory errors, differences in characteristics across architectures remain largely unexplored. To uncover these overlooked correlations, this paper conducts an extensive analysis of over 130,000 DDR4 DIMMs collected from large-scale heterogeneous production clusters over a nine-month period. Through systematic spatial and temporal analysis across two Intel x86 architectures and four major DRAM vendors, we uncover five new findings and propose a novel twostage training strategy. This strategy addresses sample quality issues by applying temporal weighting to positive samples and adaptive reweighting to negative samples. It also incorporates comprehensive multi-dimensional feature engineering, covering static, spatial, temporal, and micro-level characteristics. Finally, it integrates dual-driven sampling strategies and adaptive prediction timing to balance prediction accuracy and operational efficiency. Extensive evaluation shows that our CatBoost-based model achieves F1-scores of 49.9% on Intel x86v5 and 57.6% on Intel x86v6, substantially outperforming existing methods. This cross-architecture validation demonstrates the robustness and generalization of our approach across different hardware platforms. To the best of our knowledge, our work presents the first large-scale cross-architecture analysis of memory error patterns and provides new insights for production-scale memory failure prediction systems.

As cloud services become increasingly integral to modern IT infrastructure, ensuring hardware reliability is essential to sustain high-quality service. Memory failures pose a significant threat to overall system stability, making accurate failure prediction through the analysis of memory error logs (i.e., Correctable Errors) imperative. Existing memory failure prediction approaches have notable limitations: rule-based expert models suffer from limited generalizability and low recall rates, while automated feature extraction methods exhibit suboptimal performance. To address these limitations, we propose M2-MFP: a Multi-scale and Multi-Level Memory Failure Prediction framework designed to enhance the reliability and availability of cloud infrastructure. M2-MFP converts correctable errors (CEs) into multi-level binary matrix representations and introduces a Binary Spatial Feature Extractor (BSFE) to automatically extract high-order features at both DIMM-level and bit-level. Building upon the BSFE outputs, we develop a dual-path temporal modeling architecture: 1) a time-patch module that aggregates multi-level features within observation windows, and 2) a time-point module that employs interpretable rule-generation trees trained on bit-level patterns. Experiments on both benchmark datasets and real-world deployment show the superiority of M2-MFP as it outperforms existing state-of-the-art methods by significant margins. Code and data are available at this repository: https://github.com/hwcloud-RAS/M2-MFP.

As technology advances, streams of data can be rapidly generated in many real-life applications. This calls for stream mining, which searches for implicit, previously unknown, and potentially useful information---such as frequent patterns---that might be embedded in continuous data streams. However, most of the existing algorithms do not allow users to express the patterns to be mined according to their intentions, via the use of constraints. As a result, these unconstrained mining algorithms can yield numerous patterns that are not interesting to the users. Moreover, many existing tree-based algorithms assume that all the trees constructed during the mining process can fit into memory. While this assumption holds for many situations, there are many other situations in which it does not hold. Hence, in this paper, we develop efficient algorithms for stream mining of constrained frequent patterns in a limited memory environment. Our algorithms allow users to impose a certain focus on the mining process, discover from data streams all those frequent patterns that satisfy the user constraints, and handle situations where the available memory space is limited.