BestHub
Sign in
Tagged articles

production operations

12 articles · Page 1 of 1
Ops Community
Ops Community
Oct 28, 2026 · Backend Development

vLLM Production Tuning: The max_num_seqs, GPU Memory & Concurrency Triad

This comprehensive guide explains how vLLM's three core parameters — gpu_memory_utilization, max_model_len, and max_num_seqs — form an interdependent constraint triangle, and provides formulas, profiling log analysis, benchmarking methods, deployment configs, and five real-world troubleshooting cases to systematically tune LLM inference services for latency, throughput, and stability.

GPU memory managementKV cacheKubernetes deployment
0 likes · 67 min read
vLLM Production Tuning: The max_num_seqs, GPU Memory & Concurrency Triad
Digital Deification
Digital Deification
Oct 14, 2026 · R&D Management

FDE Four-Ring Methodology: Delivering AI Value Through Four Integrated Rings

The article details the FDE four-ring methodology—business reshaping, AI engineering, evaluation & verification, and production operations—emphasizing they form an integrated network, not sequential stages, with ontology modeling as the semantic foundation, and highlights pitfalls, deliverables, and the need for co-creation and agile iteration.

AI EngineeringAI deliveryFDE
0 likes · 16 min read
FDE Four-Ring Methodology: Delivering AI Value Through Four Integrated Rings
Architect Practice
Architect Practice
Oct 5, 2026 · Artificial Intelligence

AI Writes Code, But Misses Trade-offs: Why Software Engineering Fundamentals Are Essential

Andrew Ng argues that software engineering fundamentals are more crucial than ever for AI engineers because coding agents can implement solutions but cannot recognize critical trade-offs in latency, reliability, security, and data architecture; engineers must understand full-stack systems, data management, architecture evolution, security shift-left, and production operations to guide AI effectively.

AI EngineeringAndrew NgCoding Agents
0 likes · 15 min read
AI Writes Code, But Misses Trade-offs: Why Software Engineering Fundamentals Are Essential
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Sep 30, 2026 · Backend Development

Xiaohongshu's Agentic Release Harness: Managing Overnight Deployments at Scale

Xiaohongshu engineered a Harness system to manage agent-driven recommendation releases spanning 10+ hours across thousands of pods, using persistent tasks, declarative templates, a reconciliation engine, and effect-verified external actions to reliably execute cross-day deployments with human-in-the-loop governance.

Agentic SystemsHarness²Long-running Tasks
0 likes · 24 min read
Xiaohongshu's Agentic Release Harness: Managing Overnight Deployments at Scale
Random Bulletin
Random Bulletin
Sep 9, 2026 · Operations

Change Windows at 10M QPS: Controlling Risk, Not Just Time

This article explains why anytime deployment fails at 10M QPS scale and details a comprehensive change window mechanism that controls change quantity, risk exposure rhythm, personnel availability, and recovery resources, including layered windows, admission criteria, controlled rollout phases, freeze periods with emergency lanes, automated gates, and effectiveness metrics.

Release Engineeringautomation gatescanary deployment
0 likes · 30 min read
Change Windows at 10M QPS: Controlling Risk, Not Just Time
AI Cyberspace
AI Cyberspace
Sep 8, 2026 · Artificial Intelligence

General Agent Harness Architecture: From 4 LLM Defects to 36 Functional Modules

This article derives a universal Agent Harness architecture from four mainstream coding agents, identifying four fundamental LLM limitations — no external perception, no action capability, no memory across calls, and no guaranteed correctness — and mapping them to eight responsibility domains with 36 concrete functional modules, plus a five-dimensional framework for vertical scenario adaptation.

Action ControlCoding AgentsHarness engineering
0 likes · 53 min read
General Agent Harness Architecture: From 4 LLM Defects to 36 Functional Modules
Internet Backend Architecture Stack
Internet Backend Architecture Stack
Aug 13, 2026 · Databases

Redis Cluster Production Guide: HA, Sharding, Hot Keys & Pitfall Avoidance

This article provides a comprehensive production guide for Redis Cluster, covering cluster selection, hash slot principles, data skew solutions for big and hot keys, scaling strategies, master-slave replication latency handling, failure scenarios like split-brain, and a 7-point checklist with a real e-commerce case study.

Big KeyCaffeineData Skew
0 likes · 13 min read
Redis Cluster Production Guide: HA, Sharding, Hot Keys & Pitfall Avoidance
Xiaolin Talks Programming
Xiaolin Talks Programming
Aug 11, 2026 · Backend Development

Building Stream Processing Apps with Spring Boot & Kafka Streams: State Stores, Windowing & Exactly-Once

This article provides a comprehensive guide to integrating Apache Kafka Streams with Spring Boot for production-grade stream processing, covering core concepts, state storage with RocksDB, window aggregation strategies, exactly-once semantics, interactive queries, multi-tenancy isolation, and operational best practices.

Exactly-OnceInteractive QueriesKafka Streams
0 likes · 16 min read
Building Stream Processing Apps with Spring Boot & Kafka Streams: State Stores, Windowing & Exactly-Once
Lakehouse Research Base
Lakehouse Research Base
Jul 14, 2026 · Big Data

Flink + Paimon End-to-End Exactly-Once: Production Implementation Guide

This guide details how to achieve end-to-end exactly-once semantics with Apache Flink and Apache Paimon, covering consistency models, three-layer coordination mechanisms, production-grade configurations, concurrency rules, verification tests, failure injection scenarios, performance trade-offs, and operational SOPs for lakehouse streaming workloads.

Apache FlinkApache PaimonCheckpoint
0 likes · 26 min read
Flink + Paimon End-to-End Exactly-Once: Production Implementation Guide
Xiaolin Talks Programming
Xiaolin Talks Programming
Jun 25, 2026 · Backend Development

Building a Production-Grade Search Engine with Spring Boot & Elasticsearch

This article details how to replace MySQL complex queries with Elasticsearch in production, covering data sync via Canal/Kafka, mapping design with IK analyzer, Spring Data ES query builders, deep pagination with Search After, and consistency guarantees through reconciliation and dead-letter queues.

Consistency PatternsData SynchronizationDeep Pagination
0 likes · 21 min read
Building a Production-Grade Search Engine with Spring Boot & Elasticsearch