The paper presents a CCD‑level load‑aware thread orchestration framework that boosts vector ANNS throughput up to 3.7×, cuts P999 tail latency by 30%‑90%, reduces L3 cache miss rates by 6%‑30% and CPU stall time by 20%‑80% on AMD EPYC multi‑chiplet CPUs.
This article dives deep into Linux pthread fundamentals, covering process‑vs‑thread concepts, the POSIX API, kernel implementation via the clone syscall, thread lifecycle, synchronization primitives, common pitfalls such as deadlocks, stack overflows and thread leaks, and provides practical debugging and mitigation techniques with real code examples.
This article explains how modern multi‑core CPUs use time slices, hyper‑threading, and various context‑switching mechanisms—including preemptive and cooperative scheduling—to manage threads efficiently, and offers practical tips for reducing switching overhead and optimizing thread counts.
This article explores Java virtual threads introduced in JDK 21, detailing their architecture, implementation, practical usage with Thread and Executors APIs, performance comparisons against platform threads, best‑practice guidelines, and benchmark results demonstrating superior throughput for high‑concurrency server applications.
This article explains Java virtual threads, contrasting them with platform threads, and shows how to create, schedule, and use them effectively with Thread.Builder and Executors to achieve high‑throughput, low‑latency server applications while avoiding common pitfalls like pooling and pinning.
This article provides a comprehensive interview guide for embedded system positions, covering Linux thread scheduling, single‑core CPU execution, STM32 chip specifications, SPI communication, DMA, synchronization primitives, priority inversion solutions, debugging techniques, and memory layout fundamentals.
This article explains what threads are, how Java maps its threads to operating‑system thread models (1:1, N:1, N:M), the differences between kernel and user threads, thread scheduling, context switching, priority handling, and includes sample Java code to compare serial and multithreaded execution.
Energy Aware Scheduling (EAS) in the Android kernel uses the CPU energy model to select the most energy‑efficient CPU for each thread on heterogeneous big.LITTLE processors, considering capacity, OPP, power, and efficiency ratio; it migrates only if the migration saves more than 6% energy and overall CPU utilization stays below 80%, otherwise reverting to traditional load‑balancing.
This article explains how modern multi‑core CPUs use time slices, hyper‑threading, and various types of context switching to achieve concurrency, discusses the performance overhead of frequent switches, and offers practical guidelines for optimizing thread count and reducing switching costs.
This article explains the design of a compact RPC service that aggregates product benefit information, detailing how Sirector thread scheduling reduces CPU consumption and how abstract handlers and Spring prototype beans improve code maintainability and extensibility.
This article details Alipay's evolution from basic thread pools to a sophisticated, multi‑layered scheduling framework that balances UI responsiveness, task prioritization, and resource control across Android devices, while outlining the technical challenges and future directions of fine‑grained scheduling.
This article explores thread blocking and waking mechanisms in Java using LockSupport and Unsafe, explaining how these low-level tools enable thread scheduling without traditional locks.
RxJava brings reactive, event‑driven programming to Java by using Observables and Observers, offering operators, schedulers for thread control, and asynchronous, non‑blocking execution that avoids callback hell, improves performance, and fits UI events, I/O, RPC, and complex service orchestration, as demonstrated by Alibaba’s Xianyu.
This article explores why fast app startup matters, using Gaode Map as a case study to detail techniques such as Universal/App Links, H5 launch page handling, download size reduction, Android .so loading hooks, thread scheduling, task orchestration, and the tooling needed for performance analysis on both iOS and Android.
This article provides a comprehensive overview of process and thread scheduling in operating systems, explaining the role of the scheduler, different scheduling algorithms such as FCFS, SJF, round‑robin, priority, multilevel‑queue and lottery, their classification for batch, interactive and real‑time environments, and the key performance goals like fairness, throughput and response time.
This article provides a comprehensive overview of RxJava2 for Android, covering its observer types, backpressure strategies, thread‑scheduling transformers, the compose() operator, differences between compose and flatMap, and the distinction between hot and cold observables with practical code examples.
