Official JD Tech public account delivering best practices and technology innovation.
This article explains how to use Spring's built‑in cache annotations such as @EnableCaching, @Cacheable, @CachePut, and @CacheEvict to implement Redis‑backed caching in a Spring Boot application, covering configuration, core code, testing, and practical considerations for cache consistency and expiration.
The article presents a lightweight tool to handle data consistency in RPC services during network timeouts by mocking or spying on service calls, ensuring data integrity without actual execution.
This article presents a pioneering billion‑parameter pure time‑series large model (PCTLM) trained on a 1.5‑billion‑sample dataset, introduces a novel RLHF framework (TPO) for time‑series forecasting, and demonstrates state‑of‑the‑art performance across multiple public benchmarks, surpassing existing models such as GPT4TS.
The article explains why designing RPC interfaces with generic Result objects that contain errorCode, errorMessage and data defeats RPC's purpose, and demonstrates how returning plain business objects and leveraging Java exceptions leads to cleaner, more maintainable backend code.
The article analyzes various Java string‑replacement techniques—including simple String.replace, compiled regular expressions, Aho‑Corasick automaton, and custom Trie implementations—by presenting their designs, object sizes, and benchmark results to guide developers in choosing the most efficient solution for large keyword sets.
This article explains how to develop robust AI agents by breaking down large prompts, selecting appropriate tools, managing context efficiently, and applying modular design principles to reduce token costs, avoid hallucinations, and improve overall performance and reliability.
The article presents a self‑developed dynamic cross‑platform framework that enables a single codebase to run on iOS, Android, H5 and HarmonyOS, reducing development costs while delivering high performance, real‑time updates, and a seamless user experience across multiple devices.
This article analyzes a null pointer exception caused by InheritableThreadLocal and thread pool reuse, explaining how shared threads in a pool can lead to data corruption when parent thread values are not properly managed.
This article explains the inner workings of Caffeine's cache, covering the getIfPresent method, the design of ReadBuffer and StripedBuffer, eviction policies, the maintenance cycle, and the climb algorithm that dynamically adjusts window and protected partitions for optimal performance.
This article presents MaRCA, a multi‑agent reinforcement learning framework that allocates computation resources across the full ad‑serving chain by modeling user value, compute consumption, and action rewards, enabling fine‑grained power‑tilting toward high‑quality traffic and achieving significant business gains under strict latency constraints.
