「cpu」の検索結果

Diagnose GKE Cloud TPU training throughput drops and step-time regressions (15%+ TPU duty-cycle drop) using ML Diagnostics Workload Monitoring (`gcloud alpha mldiagnostics monitored-events` / `hypercomputecluster.googleapis.com/v1alpha`) and 1-minute Cloud Monitoring system metrics (`kubernetes.io/node/accelerator/*`). Distinguishes hardware and network fabric throttling from workload resource bottlenecks (HBM capacity, host memory, or host CPU saturation). Use when TPU training throughput or duty cycle drops without crashing pods, when `PERFORMANCE_DEGRADATION` monitored events fire, or when triaging slow multi-slice training steps. Don't use for complete multi-slice XLA execution stalls with `HANG_DETECTED` logs (use gke-ai-troubleshooting-tpu-mxla-hang) or pod eviction/interruption restarts (use gke-ai-troubleshooting-jobset-interruption).

日本語の概要は準備中です。原文の説明を表示しています。

Generates and updates secure, production-ready Kubernetes YAML manifests optimized for GKE Autopilot and GKE Standard clusters. Use when creating or modifying GKE deployment manifests, configuring container security contexts, setting CPU/memory resource limits, defining readiness/liveness/startup probes, mounting secrets and volumes, configuring GKE Gateway API routes, targeting Spot VMs, or deploying AI model inference workloads (vLLM, TGI, Gemma). Don't use for live cluster operations, pod troubleshooting (use gke-workload-troubleshooting), or cluster infrastructure provisioning (use gke-cluster-creation).

日本語の概要は準備中です。原文の説明を表示しています。

Optimizes GKE costs, rightsizes workloads, and configures Spot VMs, CUDs, cost allocation, and resource quotas. Use when optimizing GKE cluster or workload costs, configuring GKE cost allocation or quotas, rightsizing CPU/memory requests, or selecting Spot VMs and machine types. Don't use for general compute class provisioning or GPU Selection (use gke-compute-classes instead).

日本語の概要は準備中です。原文の説明を表示しています。

Configures best-practice, high-signal alerting policies for Cloud Run resources on Google Cloud (services, jobs, and worker pools) based on seasoned SRE practices. Use when analyzing, recommending, writing, or deploying Terraform PromQL alerting policies to monitor Cloud Run error rates (4xx/5xx), request latency, container instance saturation (warning/critical), container CPU/memory utilization and allocation, billable instance time, job execution status, and worker pool queue backlog. Don't use for GKE workloads (use gke-alert-configuration) or Compute Engine VMs.

日本語の概要は準備中です。原文の説明を表示しています。

MNN QNN（高通 HTP/NPU）后端的问题定位、修复与算子适配。覆盖运行时报错（1002/6000/1003/6004）、模型转换/算子校验失败、推理结果异常、精度不达标、以及性能问题;并指导为 QNN 新增/适配算子。核心方法:用 MNN2QNNModel --dump_intermediate_outputs 一次性导出 QNN 全部中间张量,与 CPU(fwd=0) 基线逐张量对比,直接定位首个出错算子(旧的 testMNNFromOnnx.py 截断二分已降为回退手段);区分“真 bug/量化/HTP fp16 精度”;新增算子先查 SDK 算子文档(MasterOpDef/HtpOpDefSupplement)再实现。

日本語の概要は準備中です。原文の説明を表示しています。

MNN Vulkan 后端 kernel/算子性能优化与新特性集成。覆盖 benchmark 基线、kernel 优化迭代、集成验证全流程，以及 GLSL .comp + makeshader 双轨、conv1x1/attention dispatcher 多路径（coopMat/subgroup/nosubgroup）、cooperative matrix、packed weight 设计、pipeline cache、CPU 侧调度瓶颈、Adreno/Mali/Apple 多 vendor 真机验证等参考知识。

日本語の概要は準備中です。原文の説明を表示しています。

Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.

日本語の概要は準備中です。原文の説明を表示しています。

Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).

日本語の概要は準備中です。原文の説明を表示しています。

Configures monitoring systems, implements structured logging pipelines, creates Prometheus/Grafana dashboards, defines alerting rules, and instruments distributed tracing. Implements Prometheus/Grafana stacks, conducts load testing, performs application profiling, and plans infrastructure capacity. Use when setting up application monitoring, adding observability to services, debugging production issues with logs/metrics/traces, running load tests with k6 or Artillery, profiling CPU/memory bottlenecks, or forecasting capacity needs.

日本語の概要は準備中です。原文の説明を表示しています。

Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving.

日本語の概要は準備中です。原文の説明を表示しています。

WPA3 / SAE (Simultaneous Authentication of Equals) attack methodology — transition-mode (mixed WPA2/WPA3) downgrade, Dragonblood side-channel attacks (CVE-2019-9494, 9495, 13377, 13456), SAE auth flooding for AP CPU exhaustion, Hash-to-Element (H2E) timing analysis, group downgrade, and 6 GHz / Wi-Fi 6E spec implications (PMF mandatory, no transition mode allowed). Use when target advertises WPA3-SAE or WPA3-Personal/Enterprise, or operates in 6 GHz where WPA3 + PMF are required by spec.

日本語の概要は準備中です。原文の説明を表示しています。

Capture and interpret iOS Simulator ETTrace profiles. Use when profiling launch or runtime latency, comparing traces, or finding CPU-heavy stacks.

日本語の概要は準備中です。原文の説明を表示しています。

Build a live Three.js landscape that stays quiet behind a subject — a noise heightfield on a polar grid so resolution follows the lens, ground coloured by slope and moisture rather than by texture, instanced GPU grass whose wind costs nothing on the CPU, scattered stones, a gradient sky dome, a star field you can actually see, and a time-of-day system that cross-fades instead of cutting. Use for hero backdrops, product stages, scroll worlds, or any scene where a building, object, or figure has to sit in a place rather than float on a gradient.

日本語の概要は準備中です。原文の説明を表示しています。

Profile, diagnose, and improve Three.js or WebGL game performance without regressing gameplay. Use for frame-time drops, CPU/GPU pressure, draw calls, texture and geometry budgets, animation loops, adaptive quality, mobile performance, and browser performance verification.

日本語の概要は準備中です。原文の説明を表示しています。

Build the effects of an ice and frost skill as one visual language in Three.js. Frost that grows like real frost, not a painted fern: a dendritic tree grown on the CPU from a seed point, stems racing outward with barbs and sub-barbs that compete for space in one time-ordered queue, side branches leaned toward their tips, baked into a distance field so the shader draws hairline feathers growing out from the seed with a bright tip, rime, crystal glints and a contact shadow. Faceted ice spikes built from half-spaces and ray-traced against their own planes (Snell, Fresnel, total internal reflection, Beer-Lambert blue, trapped bubbles, a fracture network, frosted edges) that erupt along the frost line, crack along visible planes and shatter into convex pieces that tumble, land on a face and rest, with snow lumps and ice chips settled into the crust. A low freezing mist marched as wind-stretched banks that churn in place and part around the ice, a blizzard column of powder on helical strands with a fibrous veil, glinting diamond dust, and a scanned snow floor with dunes, broken crust, a raking moon with self-shadow and two populations of glints. Use for ice or frost skills, frost novas, ice spikes or walls, glacial eruptions, shattering ice, blizzards, snow devils, freezing mist, frozen ground, "ice and frost", "frost creep", "ice shards", "realistic ice" or "skill effects only". Includes two reusable modules and a five-beat demo on a scanned CC0 snow floor with a failure-mode switch.

日本語の概要は準備中です。原文の説明を表示しています。

Profile, audit, and optimize frontend page performance with emphasis on animation work, memory-leak risks, long-session slowdowns, CSS animations, canvas/WebGL requestAnimationFrame loops, marquees, skeletons, GSAP/Three/Matter effects, timers, listeners, and observers. Use when the user asks to make animations performant, pause offscreen animations, look for memory leaks, profile pages that slow the computer over time, fix janky scrolling, reduce CPU/GPU use, or repeat the "only play in view" optimization on React/Vite/Next/frontend pages using Codex Browser.

日本語の概要は準備中です。原文の説明を表示しています。

Guide performance profiling for Apple platform apps with Instruments, Xcode diagnostics, and MetricKit. Use when investigating app hangs, stutters, high CPU, memory leaks, memory growth, OOM crashes, slow launch, battery drain, thermal issues, App Store performance readiness, or when adding os_signpost and measurement hooks.

日本語の概要は準備中です。原文の説明を表示しています。

Diagnose demonstrated CPU, command latency, allocation or IO regressions in an owned process with controlled measurements and language-appropriate profilers. Route benchmark effectiveness, score growth and trajectory efficiency to loopx-benchmark; profiling does not qualify a budget or grant execution authority.

日本語の概要は準備中です。原文の説明を表示しています。

Route C# and .NET requests to the exact installed specialist or smallest dotnet/skills marketplace plugin. USE FOR: "which specialist should own this" or "how do I add the skill" requests involving ASP.NET Core endpoints, Blazor, MAUI binding, Windows Forms specialist selection or installation, EF Core queries, test or framework migration, runtime CPU/allocation evidence, file-based C#, editor/compiler defects, a surviving MSBuild `.binlog`, or a plugin missing from `/skills`; also use for C# semantics when no narrower specialist exists. DO NOT USE FOR: requests that already name the exact installed specialist to invoke, or work unrelated to C# or .NET.

日本語の概要は準備中です。原文の説明を表示しています。

Audit and improve SwiftUI runtime performance from code review and architecture. Use for requests to diagnose slow rendering, janky scrolling, high CPU/memory usage, excessive view updates, or layout thrash in SwiftUI apps, and to provide guidance for user-run Instruments profiling when code review alone is insufficient.

日本語の概要は準備中です。原文の説明を表示しています。

Query Finelog logs and telemetry for Iris tasks, workers, profiles, training, vLLM, and cross-cluster forwarding. Use for schema discovery, SQL, memory or CPU summaries, hub-versus-regional comparisons, counter semantics, and query-performance diagnosis.

日本語の概要は準備中です。原文の説明を表示しています。

This skill should be used at the start of any computationally intensive scientific task to detect and report available system resources (CPU cores, GPUs, memory, disk space). It creates a JSON file with resource information and strategic recommendations that inform computational approach decisions such as whether to use parallel processing (joblib, multiprocessing), out-of-core computing (Dask, Zarr), GPU acceleration (PyTorch, JAX), or memory-efficient strategies. Use this skill before running analyses, training models, processing large datasets, or any task where resource constraints matter.

日本語の概要は準備中です。原文の説明を表示しています。

Cross-platform, hardware-adaptive system optimizer. Auto-detects macOS / Linux / WSL / Windows (MINGW/Cygwin/MSYS2) and CPU/RAM/disk/GPU profile, then picks the right cleanup strategy. Scans reclaimable disk space, memory pressure, runaway processes, startup bloat, network issues. CleanMyMac built into Claude Code.

日本語の概要は準備中です。原文の説明を表示しています。

Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.

日本語の概要は準備中です。原文の説明を表示しています。