Capturing thoughts and reflections while coding.
The article reveals that recent Nginx releases have made HTTP/1.1 keep‑alive the default configuration, eliminating the need for explicit proxy_http_version and Connection header settings, and explains how this reduces handshakes, lowers latency, and improves first‑byte response times for typical web applications.
Go’s database/sql package can deadlock when a Scan holds a read lock, a cancelled context triggers Close, and Columns attempts another read lock, but the new closingMutex introduced in Go’s source replaces the generic sync.RWMutex to allow reads during pending writes, preventing the deadlock.
An extensive benchmark compares Go’s standard net client with the event‑driven cloudwego/netpoll client under 60,000 concurrent connections, revealing how goroutine explosion, memory usage, and scheduler overhead differ, and demonstrates how a single scheduler plus a bounded goroutine pool dramatically reduces resource consumption.
A three‑hour benchmark on an 8C‑16G Linux host compares the standard Go net client with the netpoll client under 60,000 concurrent connections, revealing a 27.6% drop in client memory, a 99.99% cut in goroutine count, a 29.5% reduction in host memory, and a 40.7% lower CPU usage while maintaining the same throughput.
This article details a step‑by‑step guide for setting up the Qwen3‑8B large language model on a Windows 11 system using WSL2, covering hardware specs, CUDA configuration, 4‑bit quantization with BitsAndBytes, SDPA attention optimization, CPU offload, and resource‑limiting tricks to achieve smooth inference performance.
Learn how to use the Chainlit Python framework to create a web‑based Qwen3 chat interface, covering its core features, hook mechanisms for data layers, authentication, chat start, message handling, streaming generation, performance monitoring, and session restoration, with full code examples and SQLite persistence.
Through a controlled experiment with four Spring Boot service groups on Kubernetes, this article shows that relying on Java’s default GC and heap settings can drastically reduce throughput and increase tail latency, especially under higher load, and demonstrates how explicit GC algorithm and Xms/Xmx tuning restores performance.
This article explains the zero‑byte nature of Go’s empty struct, how the runtime treats zero‑size allocations, why multiple instances share the same address, and demonstrates practical patterns such as set maps, signal channels, and worker coordination that improve memory efficiency and performance.
This article explains how backend engineers should rethink table design, indexing, partitioning, and key strategies when migrating attendance data from MySQL's OLTP model to Apache Doris 2.1.7's OLAP architecture, providing concrete DDL examples and practical tips.
This article walks through the complete setup of a hybrid retrieval pipeline on two CPU‑only Linux servers using Redis 8, Qwen‑3‑Embedding vectors, and RediSearch to combine BM25 keyword scores with cosine‑based vector similarity, showing environment details, index creation, data ingestion, the hybrid_search function implementation, result normalization, and a common pitfall of forgetting to set the query language to Chinese.
