This article explains how to enable Linux CGroup support in Hadoop Yarn NodeManager to limit container CPU usage, detailing required configuration properties, hierarchy setup, CPU limit parameters, and a critical kernel version caveat.
This article provides a step‑by‑step methodology for diagnosing and resolving common online failures in Java services, covering CPU bottlenecks, disk I/O problems, memory leaks, garbage‑collection inefficiencies, and network anomalies such as timeouts, TCP queue overflows, and RST packets.
This article explains the CPU interrupt mechanism, describing how hardware devices generate interrupt signals, how the legacy 8259A PIC and modern APIC manage and prioritize those interrupts, and how interrupt affinity can be configured to improve multi‑core performance.
This article explains Linux CPU states, identifies typical MySQL queries and configurations that cause high CPU usage, and provides practical optimization techniques—including indexing, buffer tuning, and proper use of system tools—to diagnose and reduce MySQL-related CPU bottlenecks.
This article traces the evolution of Chinese domestic CPUs, examines the major manufacturers such as Feiteng, Kunpeng, Haiguang, Loongson, Zhaoxin, and Shenwei, and analyzes their architectures, performance, ecosystem, and market impact within the broader semiconductor industry.
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 detailed historical review of China’s domestic CPU industry and analyzes the strategies, product lines, performance specifications, ecosystem building, and market positioning of leading Chinese CPU manufacturers such as Phytium, Kunpeng, Haiguang, Loongson, Zhaoxin, and Shenwei.
This article compares monitoring metrics for CPU, memory, disk, and network between traditional KVM-based servers and Kubernetes pods, explaining why their indicators differ, how resource isolation works, and what key metrics users should watch to diagnose performance bottlenecks.
This guide walks through a systematic approach to troubleshooting Java service outages, covering CPU, disk, memory, GC, and network problems, and demonstrates how to use tools such as ps, top, jstack, iostat, jstat, netstat, ss, and various dump commands to pinpoint and resolve the root causes.
This article explains the essential components of a server—including power supply, CPU, RAM, storage, RAID cards, remote management modules, and motherboard—detailing their functions, typical configurations, and best‑practice considerations for reliable enterprise operations.
This guide explains how to use a variety of Linux performance tools—including top, perf, sar/ksar, mpstat, uptime, vmstat, pidstat, time, cpustat, htop, atop, glances, nmon, pcp‑gui, collectl and colplot—to monitor CPU, memory, disk, network and process activity with concrete command examples and visualisations.
This guide walks you through systematic troubleshooting of Java server incidents covering CPU, memory, disk, garbage collection, and network problems, offering step‑by‑step command‑line techniques, analysis of thread stacks, GC logs, native memory tracking, and TCP diagnostics to pinpoint root causes efficiently.
This article introduces the essential components of a server—CPU, GPU, and memory—explaining their roles, characteristics, and common configurations, especially for video processing and artificial intelligence workloads, while providing visual diagrams and further reading suggestions.
The article provides a comprehensive overview of heterogeneous computing, detailing its definition, real‑world system examples, performance advantages, key programming frameworks such as OpenCL and CUDA, industry trends like SOC integration, and a comparative analysis of CPUs, GPUs, FPGAs and ASICs.
This article explains the core components and operation of a CPU, covering the fetch‑decode‑execute cycle, instruction sets, registers, pipeline and superscalar designs, multi‑core and hyper‑threading concepts, and the hierarchy of caches from registers to L3, providing a comprehensive fundamentals overview.
This article compares CPU, memory, disk, and network monitoring metrics between traditional KVM servers and Kubernetes pods, explaining the underlying reasons for differences and offering guidance on interpreting the metrics for effective performance troubleshooting.
This article analyzes how system architects can select the optimal heterogeneous compute cores—CPU, GPU, FPGA, ASIC, VPU, or TPU—for edge AI deployments, weighing performance, size, weight, power, and cost to maximize inference efficiency and security.
Meituan’s DBA team discovered that Intel Skylake CPUs dramatically increased PAUSE‑instruction latency, causing MySQL’s ut_delay spin‑wait loops to consume most CPU cycles and reduce write throughput, and they restored performance by back‑porting MySQL 8.0’s spin_wait_pause_multiplier patch to 5.7, upgrading to CentOS 7, and moving to Cascadelake hardware.
This article explains CPU fundamentals, distinguishes physical and logical cores, shows how to retrieve CPU details on Linux, defines CPU usage and load average, offers practical thresholds, and provides step‑by‑step troubleshooting techniques for high user‑mode CPU consumption.
This article explains heterogeneous computing and compares major processing units—CPU, GPU, FPGA, ASIC, and NPU—highlighting their architectures, strengths, and typical use cases, especially in deep‑learning and AI workloads.
This article explains CPU fundamentals, the difference between physical and logical cores, how to query CPU details on Linux, the meaning of CPU utilization and load average, and provides practical steps for diagnosing high user‑mode CPU usage, especially in Java applications.
This article presents a concise, minute‑level troubleshooting workflow for Java services experiencing sudden CPU spikes and frequent Full GC, covering process inspection, thread analysis, GC monitoring, heap dump, and common root‑cause patterns such as memory overuse, CPU‑intensive code, lock misuse, and thread blocking.
After an alert showed a data platform server’s CPU usage soaring to 98.94%, this article walks through a systematic investigation—from spotting the high‑load process with top, tracing the offending Java thread using pwdx and jstack, to optimizing the time‑conversion utility that caused the overload.
Backend developers often face sudden high CPU usage or thread deadlocks in production Java services, and this article introduces a set of one‑click scripts and practical examples that automate diagnosis, locate offending code lines instantly, and provide additional utilities for classpath conflicts, JVM monitoring, and runtime tracing.
A comprehensive guide that walks you through the inner workings of a computer—from the CPU’s fetch‑decode‑execute cycle and register set, through memory hierarchy and binary arithmetic, to function calls, assembly language, and essential operating‑system concepts like APIs, multitasking and virtual memory.
Explore the essential building blocks of modern computing, covering CPU architecture, memory hierarchy, binary operations, compression techniques, operating system fundamentals, and assembly language, with clear explanations, diagrams, and code examples that demystify how hardware and software interact at the lowest level.
This article explains how to diagnose and optimize Linux system performance by examining hardware resources, OS configuration, storage setups like RAID, network bandwidth, and application code, offering a step‑by‑step guide for administrators and engineers to achieve optimal efficiency.
An unexpected CPU spike in a Java service was traced to a complex URL‑validation regex whose NFA backtracking caused catastrophic performance, and the article explains the regex engine’s behavior, identifies the problematic pattern, and shows how to refactor the expression to eliminate excessive backtracking.
The article explains the origins, metrics, and test suites of TPC and SPEC benchmarks, describes their various models for CPU, web, HPC and storage performance, shows how to query official results, and notes a promotional bundle of technical e‑books.
CPU caches, organized into L1‑L3 levels, accelerate memory access by exploiting locality, but their independent copies can cause data inconsistency across cores; coherence protocols such as MESI and memory‑barrier instructions ensure that reads and writes remain ordered and visible across all processors.
The study traced severe Redis latency spikes in an 8‑core container to the Go runtime mistakenly reading the host’s CPU count, which created hundreds of threads and scheduler contention; limiting GOMAXPROCS to the container’s quota eliminated the jitter, while further gains were suggested by reducing HMGET calls, tuning GC, and upgrading Go.
This article examines why Java applications experience high CPU usage, covering infinite loops, frequent Young GC, thread count and state, user‑space versus kernel‑space metrics, and provides a step‑by‑step troubleshooting guide illustrated with a real incident.
The article details a performance bottleneck investigation of an Erlang‑based service experiencing high CPU usage, describing the use of recon tools, pressure testing, analysis of Kafka and Nginx impacts, and the subsequent optimizations that doubled throughput to meet business requirements.
This article explains how a CPU executes programs through the fetch‑decode‑execute cycle, describes instruction sets, registers, pipelines, superscalar and multithreaded designs, and details the cache hierarchy from registers to L3, providing a comprehensive overview of modern processor fundamentals.
The article reviews the evolution of artificial‑intelligence hardware, comparing traditional CPUs with parallel GPUs, reconfigurable FPGAs, fully custom ASICs, and emerging neuromorphic chips, highlighting their architectures, performance trade‑offs, power consumption, and current industry adoption.
The article explains how CPUs and GPUs differ in architecture and workload handling, details key GPU specifications such as CUDA cores, memory bandwidth and floating‑point precision, reviews NVIDIA's product families and architectural evolution, and highlights the role of GPUs in deep learning training and inference while also mentioning a related technical ebook promotion.
This article explains how a CPU executes programs through the fetch‑decode‑execute cycle, describes instruction sets, registers, pipelines, superscalar and multithreaded designs, and details the hierarchy of caches from registers up to L3, providing a comprehensive overview of modern processor fundamentals.
This article compares various AI server architectures—CPU, GPU, FPGA, TPU, and ASIC—by evaluating performance versus programmability, and outlines practical guidelines for choosing GPU servers based on workload, cost, power, and deployment scenarios.
This article explains how basic semiconductor concepts and diode‑based logic gates can be combined into simple arithmetic circuits, registers, and a minimal instruction set to construct a functional 4‑bit CPU, illustrating each step with diagrams and historical context.
The article explains the structural differences between CPUs and GPUs, their respective design goals, and why GPUs excel at parallel image and AI workloads, while also noting a New Year promotional bundle of technical e‑books priced at 168 yuan.
This article summarizes the latest 2020 processor announcements from ARM, Intel, and AMD, highlighting ARM's Cortex‑A77 and Mali‑G77 designs, Intel's 10 nm Ice Lake and Athena initiatives for thin laptops, and AMD's Ryzen 3000 7 nm lineup, while comparing their differing strategies for the future of computing.
This guide explains how to obtain, calculate, and interpret key Android performance metrics—including CPU usage, memory consumption, frame rate, network traffic, thread activity, and power consumption—using adb commands, code snippets, and best‑practice recommendations.
A Java service suffered near‑100% CPU usage because a URL‑validation regular expression caused catastrophic backtracking, and the article explains the NFA engine behavior, identifies the regex flaws, and shows how to rewrite the pattern with possessive quantifiers to dramatically improve performance.
The article provides a comprehensive overview of RISC principles and compares major processor families—including ARM, x86, Atom, MIPS, and PowerPC—detailing their architectures, performance characteristics, market adoption, and licensing models while also mentioning related DSP technologies and reference resources.
The article explains how modern servers with multiple CPUs and memory modules connect via legacy Front Side Bus and modern NUMA architectures, demonstrates Linux commands to inspect memory topology, and presents benchmark results showing latency differences between intra‑node and inter‑node memory accesses.
This article explains the core principles of CPU operation, covering the fetch‑decode‑execute cycle, instruction sets, registers, pipeline and superscalar designs, multithreading and multicore behavior, as well as cache hierarchy from registers through L1‑L3, illustrating how these mechanisms affect program execution.
This article explains how Linux reports CPU usage with tools like top, the meaning of the fields in /proc/stat, how utilization percentages are calculated, the concepts of run queues, load average, context switching, multi‑core scheduling, and how to use perf and taskset for deeper performance analysis.
The article surveys the evolution of artificial‑intelligence chips, comparing traditional CPU architectures with parallel accelerators such as GPUs and FPGAs, fully custom ASICs, and emerging neuromorphic chips, highlighting their structures, performance trade‑offs, and application scenarios.
This article examines three common Redis performance pitfalls—short-lived connections, frequent INFO commands, and misuse of pipelines—by presenting real‑world experiments, profiling data, and source‑level analysis, then offers concrete code changes and best‑practice recommendations to reduce CPU and memory overhead.
An intriguing experiment shows that replacing a harmless "y += 0" with "y += 0.1f" makes a tight C++ loop run seven times faster, revealing how denormalized floating‑point numbers trigger costly CPU paths and how understanding IEEE‑754 representation can restore performance.
The article explains how heterogeneous parallel computing—distributing tasks across CPUs, GPUs, FPGAs and other accelerators—has become essential after Moore’s law plateau, detailing its principles, hardware and software perspectives, classification of architectures, processing stages, user‑guided versus compiler‑guided methods, and its relevance to AI, cloud and industry workloads.
CPU and GPU serve distinct roles in computing: the CPU, as a versatile general‑purpose processor, handles complex logic and varied data types, while the GPU, built with many simple cores and long pipelines, excels at parallel processing of uniform, large‑scale data such as graphics and AI workloads.
This article explains the fundamentals of CPUs, covering their core functions, internal components, operation stages, key performance metrics such as clock speed, multiplier and external frequency, as well as modern technologies like Turbo Boost, multi‑core, hyper‑threading, and the evolution of CISC and RISC architectures.
This article explains the core principles of CPU operation, covering instruction fetching, decoding, execution cycles, register types, pipeline and superscalar architectures, multi‑core and hyper‑threading designs, as well as the hierarchy of caches from registers to L3, providing a comprehensive overview of modern processor fundamentals.
This guide explains the environment setup, connection steps, parameter selection, data collection, and result interpretation for an Android performance monitoring tool that measures traffic, memory, CPU usage, FPS, and network speed during app testing.
This article chronicles the development of the x86 and ARM ecosystems, detailing early competitors, strategic alliances, technological shifts, market battles, and future trends that have shaped the semiconductor industry and the broader computing landscape.
The article analyzes a MySQL 5.6 high‑CPU issue triggered by a frequently executed query with many UID values, identifies mutex contention and excessive memory allocation in InnoDB as root causes, and presents a solution of switching the memory allocator to tcmalloc with configuration changes.
This article walks Java developers through diagnosing and fixing common problems such as deadlocks and CPU usage spikes, explaining the underlying concepts, showing practical code examples, and demonstrating how to use tools like jps, jstack, JConsole, JVisualVM, and Arthas to pinpoint and prevent performance bottlenecks.
The article examines the recent loss of X86 licensing for Chinese chip makers, detailing AMD's withdrawal of support for Tianjin Haiguang, the expiration of VIA's agreements, and the broader challenges facing domestic X86 processors amid shifting market dynamics and emerging alternatives.
This article provides a comprehensive, illustrated guide to the entire manufacturing workflow of an Intel Core i7 CPU—from raw silicon extraction and purification, through photolithography, doping, and multilayer metal deposition, to testing, packaging, and future architectural outlooks including x86 limitations and the rise of RISC‑V and the RIOS lab.
This article debunks the myth that larger database connection pools always improve performance, explains how CPU, disk, and network I/O affect optimal pool size, presents real‑world test results, and provides a practical formula for calculating the right number of connections.
This guide walks you through practical Linux performance troubleshooting steps, covering how to inspect CPU details, list boot‑time services, monitor CPU load, check free memory, and analyze disk I/O, with concrete commands and solutions to improve system speed.
This guide walks you through essential Linux performance diagnostics—examining CPU details, startup services, load averages, memory usage, and disk I/O—providing commands and practical solutions to identify bottlenecks and improve system responsiveness for your workstation.
This article reproduces a Linux 4.18 kernel scenario to analyze why CPUs remain in C0 under power‑saving mode, examines the role of the tuned service and hyper‑threading, and provides concrete BIOS and tuned‑service configuration recommendations for different workload profiles.
This article explains how to identify and resolve common causes of Java service slowdown—such as excessive Full GC, high CPU usage, intermittent latency, waiting threads, and deadlocks—by using tools like top, jstat, jstack, and MAT for systematic troubleshooting.
This article explains how Linux kernels 2.6 and 4.18 manage CPU P‑states and C‑states, covering BIOS control, cpupower usage, idle drivers, the menu governor, MONITOR/MWAIT effects, and performance comparisons between power‑saving and performance modes on Intel Xeon E5‑2630 v4.
The article explains how a complex Java regular‑expression used for URL validation triggers catastrophic backtracking, causing near‑100% CPU usage, and demonstrates how to analyze, reproduce, and fix the regex by reducing greedy patterns and adding missing characters.
This article chronicles the development, competition, and strategic shifts of the x86 and ARM processor ecosystems from their early days through recent innovations, highlighting key architectural decisions, market dynamics, and the evolving role of software abstraction layers in shaping the semiconductor landscape.
This article walks through the complete journey of a single Java statement—from high‑level source code, through compilation to bytecode, JVM interpretation and JIT compilation, down to CPU instruction fetching, decoding, execution, caching, memory hierarchy, threading, interrupts, and the underlying Linux process model.
This article explains the complete execution path of a single Java line—from the Von Neumann CPU components, instruction fetch‑decode‑execute pipeline, Java bytecode generation, JVM class loading and interpretation, memory layout and caching, to Linux process memory management, thread scheduling, synchronization mechanisms and timer implementation—providing a deep technical foundation for Java performance tuning.
This article explains the fundamental differences between CPUs and GPUs, outlines key GPU specifications such as CUDA cores, memory capacity, bandwidth, and floating‑point precision, and reviews NVIDIA's major GPU series and architectural evolution for high‑performance and AI workloads.
This guide walks you through essential Linux performance diagnostics—checking CPU details, boot‑time services, CPU load, free memory, and disk I/O—providing concrete commands, interpretation tips, and practical solutions to identify and resolve common slowdown causes.
This article explains the fundamental architectural differences between CPUs and GPUs, their design goals and performance characteristics, and compares OpenCL and CUDA, highlighting OpenCL’s cross‑platform flexibility versus CUDA’s NVIDIA‑specific optimization, while illustrating how each fits various parallel computing tasks.
This article explains how infinite loops, frequent Young GC, thread count, and thread states affect CPU utilization in Java, describes the relationship between CPU, processes, and threads, and shares practical troubleshooting steps and real‑world cases for diagnosing high CPU usage.
This article explains the hidden multithreading bugs in JDK 7 HashMap and JDK 8 ConcurrentHashMap that can cause 100% CPU usage, demonstrates how to reproduce them, and offers practical ways to avoid or resolve the issue.
This article explores why Java applications experience high CPU usage, examines common misconceptions about loops, GC, thread counts, and blocked states, explains how CPU usage is calculated, and provides practical troubleshooting steps with real‑world examples and monitoring tips.
When a data platform server spikes to 98% CPU despite low traffic, this guide walks you through using Linux tools and Java thread analysis to locate the offending process, identify the problematic code, and apply a targeted fix that drops CPU load dramatically.
To alleviate GPU scarcity, iQIYI’s cloud platform migrated deep‑learning inference to CPUs and applied system‑level (MKL‑DNN, OpenVINO), application‑level, and algorithm‑level optimizations—tuning threads, batch size, NUMA, pruning and quantization—delivering 1‑9× speedups across thousands of cores while preserving latency and accuracy.
This guide walks through diagnosing a server’s 98% CPU spike by locating the offending process, pinpointing the problematic Java thread, analyzing the root‑cause time‑utility method, and applying a code simplification that reduces load by over 30‑fold.
The article reviews the two main development routes for artificial intelligence chips—traditional architectures that enhance CPU, GPU, and FPGA performance and emerging neuromorphic designs like IBM TrueNorth—detailing their structures, advantages, limitations, and industry adoption.
The 52nd TOP500 list released in November 2018 shows US DOE machines dominating the top ten, highlights performance gains of Summit and Sierra, details the hardware of leading supercomputers worldwide, and analyzes country shares, manufacturers, accelerator usage, interconnect technologies, and energy‑efficiency rankings such as Green500 and HPCG.
This article details a step‑by‑step investigation of frequent full GC events and high CPU usage in a Java backend service, covering memory analysis with SGM, heap dumps, identification of large Base64 image strings, static encryption utilities, and remediation actions such as heap size increase and custom encryption handling.
This article walks through setting up a Vagrant‑based experiment that injects an intentional infinite loop into a simple Spring MVC service, then demonstrates step‑by‑step how to identify the offending process using top, examine JVM heap with jstat, and trace the problematic thread with jstack to resolve a CPU‑100% issue.
The article offers a comprehensive, historically grounded analysis of the x86 and ARM ecosystems, tracing their origins, competitive battles, technical innovations, market shifts, and future challenges while highlighting why software ecosystems, licensing models, and emerging technologies shape CPU dominance.
This article walks through a practical Linux performance investigation, showing how to use the top command to pinpoint high‑CPU processes, examine thread details, convert thread IDs, analyze thread dumps for lock contention, and interpret key top output fields for effective troubleshooting.
An Alibaba tech support engineer investigates a server whose CPU appeared fully utilized, discovers the misleading 'ni' metric, reproduces the issue with high‑nice loops, uncovers hidden mining processes via ld.so.preload, and explains how to detect and mitigate such stealthy CPU hogs.
After moving a Git server to a new data center, CPU sys time surged due to thousands of sshd processes contending on XFS read‑write locks while repeatedly reading a massive /var/log/btmp file caused by PAM postlogin, and the analysis shows how perf, strace and log rotation can resolve the issue.
The article explains how heterogeneous parallel computing—leveraging CPUs, GPUs, FPGAs and other specialized units—addresses the performance limits of traditional serial programming by distributing tasks across diverse hardware, detailing its concepts, architectures, development models, and relevance to AI and cloud workloads.
This guide explains the three main performance bottlenecks—CPU, network I/O, and disk I/O—on Linux servers, offers concrete optimization techniques for each, and introduces essential diagnostic tools such as top, free, vmstat, strace, tcpdump, and gprof to pinpoint and resolve issues.
This article introduces a self‑developed Android performance monitoring app, explains how to install, start, and view test results, and details the underlying data‑collection mechanisms for memory, battery, CPU, and FPS metrics, providing practical guidance for developers.
This article examines common memory leak and performance problems in C/C++ applications, categorizing causes such as pointer loss, improper memory release, excessive fragmentation, infinite allocation, and container misuse, and discusses diagnostic techniques, underlying allocation mechanisms, and mitigation strategies.
The article explains how CPUs, despite being usable for early cryptocurrency mining, are inefficient due to limited parallelism and general‑purpose design, while GPUs and ASICs provide massive parallel integer processing that makes them far more effective for modern proof‑of‑work algorithms.
This article translates hardware latency into human‑scale time, revealing how disks, networks, memory and other components appear painfully slow from a CPU’s viewpoint and why performance optimization is a complex, system‑wide challenge.
This article walks through a real‑world incident where a data platform’s CPU usage surged to 98%, detailing how to pinpoint the high‑load process, trace the offending Java thread, uncover a time‑utility method bottleneck, and apply a concise fix that reduced load by thirtyfold.
This article translates hardware latencies—from CPU cache and memory to disk and network—into human‑scale time equivalents, using vivid analogies and benchmark data to illustrate why storage and networking appear extremely slow from the processor's viewpoint.
This article introduces the fundamentals of x86 assembly language, explaining why low‑level code is needed, the history of assembly, CPU registers, the heap and stack memory models, and walks through a complete example with detailed explanations of each assembly instruction.
This article explores Hadoop YARN’s evolution, multi‑tenant design, queue and node‑label scheduling, real‑world resource allocation challenges, and data‑driven tools that automate diagnostics and visualizations to optimize CPU and memory usage across massive clusters.
This guide walks through a systematic troubleshooting workflow for a Linux gateway that spikes to 891% CPU, using top to locate the offending process, inspecting thread usage, capturing stack traces with gstack, dumping core files, tracing system calls with strace, and analyzing the core dump in gdb to pinpoint the poll() call causing the load.
This article explains how dynamic rate limiting can automatically compute throttling thresholds using system load metrics such as CPU usage, load average, and response time, and demonstrates its effectiveness through real‑world testing on a flight‑price service.
This article introduces four Python CPU profiling tools—cProfile, line_profiler, pprofile, and vprof—explaining their installation, usage on CPython and PyPy, interpreting their outputs, and visualizing results with graphviz, helping developers identify performance bottlenecks such as costly list.append calls.
