This paper presents a cloud-native Lakehouse architecture designed for real-time semiconductor analytics, with a focus on optimizing storage tiering, data lineage, and cost-energy co-optimization. As semiconductor data analytics require processing massive amounts of real-time data, traditional data warehouses are often insufficient in addressing the need for low-latency, high-concurrency queries. The proposed framework leverages cloud-native technologies, such as AWS, Azure, and distributed databases like Apache Doris, to design a dynamic multi-tier storage system that segregates data based on access frequency and volatility, incorporating columnar compression techniques for efficient storag... More >

Yield optimization in advanced manufacturing rarely proceeds as a tidy pipeline; it arises from the gradual convergence of evidence across spatial wafer patterns, multivariate metrology, and asynchronous process and equipment events that interact in ways that are only partially observable. Prior studies often separate these modalities, assigning convolutional encoders to wafer maps, sequence models to metrology, and template based encoders to logs, an arrangement that can perform well locally yet struggles to sustain cross-modal alignment or to reason over the hierarchy that links defects to steps and equipment. Building on these observations, we introduce a manufacturing semantics oriented... More >