Results for “fault tolerance”

Knowledge base from the NASA Fault Management Handbook (NASA-HDBK-1002, Draft 2, April 2012). Use for the systems-engineering discipline of designing what a system does when it fails — the fault/failure/anomaly vocabulary, the five FM strategies (two prevention strategies — design-time fault avoidance and operational failure avoidance — plus three tolerance strategies — failure masking, failure recovery, and goal change), the six FM process activities across NASA mission phases Pre-A to E, response-latency-versus-time-to-criticality (TTC) design trades, fault/failure containment regions and FEPPs, the four redundancy approaches, top-down FM requirements development, FM verification vs. validation, the seven dedicated FM milestone reviews (FMCR/FMARR/FMPDR/FMCDR/FMTRR/FMLRR/FMCERR), FM organizational structure, and mined NASA Lessons Learned. Space-mission oriented (flight/ground/operations); spacecraft examples throughout. IMPORTANT: built from an UNAPPROVED DRAFT — several sections (detailed architecture building blocks, the Assessment & Analysis guidance, and Operations & Maintenance) are placeholders in the source and are thin here. Thin on detailed FDIR algorithms, terrestrial/IT fault management, and any approved-standard terminology that may have changed since Draft 2.

The English summary is being prepared. Showing the original description.

Adapt machine learning training to run across multiple GPUs or machines with Ray Train. Covers changes to existing code, comparing training settings with Ray Tune, and saving checkpoints so training can resume after failures.

• Adapting existing PyTorch code for multiple GPUs
• Training HuggingFace models across multiple machines
• Comparing learning rates and batch sizes with Ray Tune

Helps adapt machine learning training to multiple GPUs and machines with Ray Train. Covers resource allocation, recovery from saved training states, and Ray Tune searches that compare settings such as learning rate and batch size.

• Adapt existing PyTorch training to multiple GPUs
• Run HuggingFace training across multiple machines
• Compare learning rates and batch sizes with Ray Tune

Resilience testing specialist for failure injection, game day planning, and building confidence in system reliabilityUse when "chaos engineering, resilience testing, failure injection, game day, fault tolerance, chaos experiment, disaster recovery, reliability testing, chaos-engineering, resilience, failure-injection, game-day, fault-tolerance, reliability, testing, litmus, chaos-monkey, ml-memory" mentioned.

The English summary is being prepared. Showing the original description.

Foundational distributed computing algorithms for consensus, coordination, and fault tolerance

The English summary is being prepared. Showing the original description.

Foundational distributed computing algorithms for consensus, coordination, and fault tolerance

The English summary is being prepared. Showing the original description.

Implement circuit breaker logic for agentic tool calls — tracking tool health, transitioning between closed/open/half-open states, reducing task scope when tools fail, routing to alternatives via capability maps, and enforcing failure budgets to prevent error accumulation. Separates orchestration (deciding what to attempt) from execution (calling tools), following the expeditor pattern. Use when building agents that depend on multiple tools with varying reliability, designing fault-tolerant agentic workflows, recovering gracefully from tool outages mid-task, or hardening existing agents against cascading tool failures.

The English summary is being prepared. Showing the original description.

Foundational impossibility result proving consensus cannot be guaranteed in asynchronous systems with even one faulty process

The English summary is being prepared. Showing the original description.

Foundational impossibility result proving consensus cannot be guaranteed in asynchronous systems with even one faulty process

The English summary is being prepared. Showing the original description.

Chaos Engineering discipline — proactive experimentation on distributed systems to build confidence in their resilience. Covers Netflix-inspired chaos principles, experiment design, blast radius control, LitmusChaos and Chaos Mesh usage, game day planning, and production safety. USE WHEN: validating system resilience before production incidents, designing chaos experiments, planning game days, implementing CI/CD chaos pipeline, or building confidence in fault tolerance. Triggers on "chaos engineering", "chaos experiment", "game day", "resilience test", "fault injection", "Litmus", "Chaos Mesh", "break things on purpose".

The English summary is being prepared. Showing the original description.

Help design quorum management for distributed systems, where multiple nodes must agree on decisions. Covers adapting participation thresholds and voting weights to network conditions, performance, and failures, plus membership changes, validation, and rollback.

• Review quorum requirements under changing network conditions
• Plan node additions, removals, and failure handling
• Design weighted voting based on node capabilities

Benchmark distributed consensus protocols, methods computers use to agree on shared results. Compare processing rates, response times, scaling behavior, and resource usage across Byzantine, Raft, and Gossip, then identify bottlenecks and potential tuning changes.

• Comparing Byzantine, Raft, and Gossip performance
• Measuring response times under increasing load
• Finding CPU, memory, or network bottlenecks

Write more reliable Bash scripts for production automation, CI/CD pipelines, and system administration. Covers input validation, error handling, structured logs, safe temporary files, dry runs to preview changes, and scripts designed for safe reruns.

• Writing production automation scripts
• Improving error handling in CI/CD scripts
• Adding dry runs to deployment scripts

Evaluate Google Cloud workloads and develop reliability recommendations covering user experience goals, redundancy, monitoring, and recovery. The skill applies the Google Cloud Well-Architected Framework to help identify gaps and plan improvements suited to business requirements.

• Reviewing a Google Cloud architecture for reliability gaps
• Setting reliability targets based on user experience
• Planning redundancy and behavior during partial failures

Plan and troubleshoot large language and vision model training and inference. This guide uses hardware measurements to identify bottlenecks, address memory shortages and unstable training, and improve checkpointing, recovery, and performance across multiple GPUs and machines.

• Diagnosing slow training and GPU memory shortages
• Investigating loss spikes and invalid numerical values
• Debugging communication hangs across GPUs and machines

Resiliency features in Megatron Bridge including fault tolerance, straggler detection, in-process restart, preemption, and re-run state machine.

The English summary is being prepared. Showing the original description.

Implement reliability patterns for Claude API: circuit breakers, graceful degradation, idempotency, and fallback strategies. Trigger with phrases like "anthropic reliability", "claude circuit breaker", "claude fallback", "anthropic fault tolerance".

The English summary is being prepared. Showing the original description.

Use when oTP behaviors including gen_server for stateful processes, gen_statem for state machines, supervisors for fault tolerance, gen_event for event handling, and building robust, production-ready Erlang applications with proven patterns.

The English summary is being prepared. Showing the original description.

Multi-protocol consensus for agent swarms supporting Raft leader election, Byzantine fault tolerance, Gossip state propagation, and CRDT conflict-free merging.

The English summary is being prepared. Showing the original description.

Write idiomatic Elixir code with OTP patterns, supervision trees, and Phoenix LiveView. Masters concurrency, fault tolerance, and distributed systems. Use PROACTIVELY for Elixir refactoring, OTP design, or complex BEAM optimizations.

The English summary is being prepared. Showing the original description.

Master defensive Bash programming techniques for production-grade scripts. Use when writing robust shell scripts, CI/CD pipelines, or system utilities requiring fault tolerance and safety.

The English summary is being prepared. Showing the original description.

Multi-agent orchestration: lane contracts, worker-model doctrine, fault tolerance. Triggers: orchestrate, coordinate, agents, parallel, spawn, contract, tier 2, tier 3.

The English summary is being prepared. Showing the original description.

Post-deploy monitoring. Watches live app via headless browser for console errors, performance regressions, page failures, visual anomalies. Alert on deltas not absolutes, transient tolerance (2+ consecutive checks). Default 10-minute monitoring window. Baseline capture mode.

The English summary is being prepared. Showing the original description.

Master defensive Bash programming techniques for production-grade scripts. Use when writing robust shell scripts, CI/CD pipelines, or system utilities requiring fault tolerance and safety.

The English summary is being prepared. Showing the original description.