Best of Performance — April 2025

  1. 1
    Article
    Avatar of communityCommunity Picks·52w

    Why Companies Are Saying Goodbye to Next.js 🚪💨

    Developers and companies are increasingly moving away from Next.js due to issues like performance bottlenecks, overengineering, and vendor lock-in with Vercel. While Next.js offers benefits such as fast performance and SEO-friendly architecture, its complexities and limitations have made alternatives like Astro, Remix, and Qwik more appealing. For simpler, faster solutions, some developers are resorting to using React with Vite instead.

  2. 2
    Article
    Avatar of communityCommunity Picks·1y

    TanStack Pacer

    TanStack Pacer offers utilities for rate limiting, throttling, and debouncing, enabling developers to effectively manage application timing. It supports intelligent queuing and concurrency controls while providing pause, resume, and cancel options for complex asynchronous workflows. The API is modular and framework-independent, ensuring ease of use.

  3. 3
    Article
    Avatar of architectureweeklyArchitecture Weekly·51w

    PostgreSQL JSONB - Powerful Storage for Semi-Structured Data

    PostgreSQL JSONB offers a powerful method for storing and querying semi-structured data. JSONB stores data in a binary format, allowing for efficient indexing and querying without the need for repeated parsing. The article discusses the internal structure of JSONB, performance benefits, and various indexing strategies like GIN and B-Tree to optimize JSONB queries. It also covers practical applications, such as using JSONB for managing heterogeneous data models and ensuring query performance. Although JSONB provides flexibility, it's essential to understand its performance implications for large datasets.

  4. 4
    Article
    Avatar of things_to_knowThings To Know·1y

    Stop Overusing useEffect

    Check out this guide on when you might not need useEffect in your React components. Simplifying code and improving performance can be achieved by understanding alternative approaches.

  5. 5
    Video
    Avatar of t3dotggTheo - t3․gg·1y

    Why is Next.js so slow??

    Server components in Next.js are under scrutiny for their performance, and while some apps using server components feel slow, the issue often lies with developer implementation rather than the technology itself. The post discusses the balance between client-side and server-side rendering, emphasizing the need for proper loading states and understanding of server behavior to optimize performance. Tools like Savala can streamline deployment by managing infrastructure needs such as CDNs and databases effectively.

  6. 6
    Video
    Avatar of programmersarealsohumanProgrammers are also human·1y

    Interview with Vibe Coder in 2025

    Discussion covers the development and performance of an Uber simulator, database and web security, optimization techniques, and various app functionalities. Emphasizes a light-hearted approach to troubleshooting and iterative development, reflecting on modern app trends and developer challenges.

  7. 7
    Article
    Avatar of rubylaRUBYLAND·52w

    Fix N+1 Queries Without Eager Loading Using SQL Subquery

    Learn how to address the N+1 query problem in Rails without resorting to eager loading. By leveraging SQL subqueries, you can fetch specific associated records directly within your main query, reducing memory consumption and improving performance. The post walks through a real-world example of fetching a member's latest activity timestamp efficiently, discussing various approaches and their trade-offs before demonstrating the optimal SQL subquery solution.

  8. 8
    Article
    Avatar of portkeyportkey·52w

    Task-Based LLM Routing: Optimizing LLM Performance for the Right Job

    Task-based LLM routing directs incoming AI requests to the most suitable large language model based on the task. This approach improves performance, reduces costs, and enhances scalability by matching tasks with models optimized for those specific needs. For instance, simpler tasks can be routed to lightweight models like GPT-3.5 to minimize costs, while complex tasks are handled by more powerful models like GPT-4. This method also enhances reliability and latency, and is useful in diverse applications like customer support, content creation, code-related tasks, and multilingual processing.

  9. 9
    Article
    Avatar of dhhDavid Heinemeier Hansson·52w

    The new Framework 13 HX370

    The new AMD HX370 option in the Framework 13 laptop offers a notable performance boost for developers, running test suites significantly faster than other models. Despite its higher cost, it appeals with its better keyboard, matte screen, repairability, and excellent Linux compatibility. Alternatives like the budget-friendly Ryzen AI 5 340 and the slightly less powerful but cheaper Ryzen AI 7 350 also present competitive options. Apple's hardware remains strong, but its software is criticized for increasing bloat and control, making Linux a more attractive choice for development.

  10. 10
    Article
    Avatar of collectionsCollections·52w

    Laravel Solves the N+1 Problem Efficiently

    Laravel version 12.8 offers a new feature to efficiently handle the N+1 problem by automatically eagerly loading related data. This enhancement reduces the number of queries required, minimizes performance bottlenecks, and simplifies the development process.

  11. 11
    Article
    Avatar of hnHacker News·1y

    ferronweb/ferron: A fast, memory-safe web server written in Rust.

    Ferron is a high-performance, memory-safe web server built with Rust. It uses Rust’s async capabilities for speed, focuses on security and robust concurrency, and features a modular architecture for customization. Currently, Ferron is under development with installation instructions to be provided upon the initial release. Users can clone the repository and build the project using Cargo.

  12. 12
    Article
    Avatar of wundergraphWunderGraph·1y

    Golang sync.Pool is not a silver bullet

    Golang's sync.Pool is a thread-safe object pooling mechanism that helps reduce memory allocations and garbage collection pressure. However, it comes with potential pitfalls like unpredictable memory growth, size distribution issues, and added complexity. It is valuable in scenarios with predictable object sizes, high-frequency allocations, and short-lived objects, but should be avoided in cases with varying object sizes, low allocation frequency, and long-lived objects. Alternative approaches like direct allocation, fixed-size buffers, and multiple pools may be more appropriate for some situations.

  13. 13
    Article
    Avatar of last9Last9·1y

    HAProxy vs NGINX Performance: A Comprehensive Analysis

    Choosing between HAProxy and NGINX for load balancing is critical for system reliability and performance. HAProxy is designed specifically for load balancing with an event-driven model, while NGINX evolved from a web server to include load balancing. HAProxy excels in connection management, higher connection concurrency, and memory efficiency under high loads, making it suitable for microservices and environments with unpredictable traffic spikes. NGINX performs better at lower concurrency levels, content delivery, and SSL processing. Detailed metrics, configuration strategies, and optimization guidelines are provided for both solutions to help in decision-making and performance tuning.

  14. 14
    Article
    Avatar of awegoAwesome Go·51w

    Optimizing Heap Allocations in Golang: A Case Study

    The post discusses a 30% performance regression in a Golang codebase due to a commit that was supposed to be a no-op refactor. It highlights the importance of understanding memory allocations in Golang, particularly the difference between stack and heap allocations. The change of using a pointer receiver instead of a value receiver to prevent unnecessary copies led to performance optimization. The post also explains escape analysis and how tools like `go build -gcflags "-m"` can help in identifying allocation issues.

  15. 15
    Article
    Avatar of phProduct Hunt·50w

    RightNow AI 'V2.0' - Vibe Coding for CUDA Engineers

    RightNow AI V2.0 helps CUDA engineers automatically profile, identify bottlenecks, and optimize CUDA kernels for enhanced performance. It aims to streamline the coding process for engineers working with parallel computing capabilities.

  16. 16
    Video
    Avatar of t3dotggTheo - t3․gg·51w

    JavaScript is fast enough (don't rewrite your code)

    The discussion focuses on whether rewriting JavaScript tools in faster languages like Rust or Go is beneficial. It argues that while new languages may offer performance improvements, optimizing JavaScript itself and leveraging its ecosystem might be more practical. The piece also highlights how JavaScript's widespread use lowers the barrier for contributions, making it a more versatile and accessible language for developers.

  17. 17
    Article
    Avatar of csharpcornerC# Corner·52w

    .NET 9 Apps Are Faster Than Ever

    Microsoft’s .NET 9 introduces significant performance improvements such as enhanced JIT compilation, faster JSON serialization, reduced garbage collection pauses, and better HTTP/3 handling. Benchmark comparisons between .NET 8 and .NET 9 show substantial gains in speed, reduced memory usage, and lower CPU usage under load, making applications faster and more efficient with minimal code changes.

  18. 18
    Article
    Avatar of lobstersLobsters·52w

    Two Years of Rust

    Reflections on two years of professional experience with Rust, highlighting its strengths in performance, tooling, type safety, error handling, and productivity. Discusses both positive aspects like the borrow checker, async handling, and confidence in code, and negatives such as the module system and build performance. The author shares insights into hiring Rust developers and the emotional impact of working with the language.

  19. 19
    Video
    Avatar of dreamsofcodeDreams of Code·50w

    There's more than one way to scale Redis/Valkey to 1M op/s...

    Exploring various strategies to scale Redis or Valkyrie to handle 1 million requests per second, including vertical scaling, horizontal scaling, pipelining, read replication, and using Redis cluster. The post discusses the challenges and benefits of each approach, and introduces Dragonfly DB as a high-performance alternative that can handle requests more efficiently without complex configurations.

  20. 20
    Article
    Avatar of dockyardDockYard·52w

    Zero-Cost Abstractions in Rust: Power Without the Price

    Rust allows developers to write high-level, clean code without sacrificing performance thanks to zero-cost abstractions. This feature enables tools like loops, functions, and data transformations to run as efficiently as low-level code, providing both speed and safety. Key examples include iterators, closures, and smart pointers. These abstractions are optimized at compile time, enabling Rust to deliver exceptional performance in various applications such as REST APIs, data processing, and WebSocket handlers.

  21. 21
    Article
    Avatar of wordpresscoreMake WordPress Core·52w

    WP_Query changes in WordPress 6.8

    WordPress 6.8 introduces caching optimizations in the WP_Query::get() method and the WP_Query::$query_vars property to enhance performance, especially for sites that do not use persistent caching. These changes may impact how themes and plugins handle queries. Developers are advised to check code compatibility and use specific methods for array comparison due to these updates.

  22. 22
    Video
    Avatar of nickchapsasNick Chapsas·1y

    The "in" keyword in C# is Awesome

    The 'in' keyword in C# allows structs to be passed by reference while making them read-only within the method for performance optimization. This avoids costly copying of large structs. However, if the struct isn't marked as readonly, hidden defensive copies may occur, reducing performance benefits. Combining 'in' with 'readonly struct' or read-only members maximizes performance gains.

  23. 23
    Video
    Avatar of codeheimcodeHeim·1y

    #84 Golang - Concurrency in Go Web Applications: From net/http to Background Workers

    Discover how concurrency works in Go web applications, focusing on the use of goroutines and channels. Learn about the concurrency model in Go's net/http and Gin frameworks, and understand best practices for handling background tasks efficiently. Explore the advantages of using established background processing systems like Factory, Async, or RabbitMQ for managing heavy operations.

  24. 24
    Article
    Avatar of jetbrainsJetBrains·1y

    The .NET Tools Blog

    C# offers a variety of data structures designed for different use cases. The choice of the right collection depends on factors such as the type of operations, performance considerations, and the structure of the data.

  25. 25
    Article
    Avatar of hnHacker News·1y

    Rust-GPU/Rust-CUDA: Ecosystem of libraries and tools for writing and executing fast GPU code fully in Rust.

    The Rust CUDA Project aims to make Rust a tier-1 language for high-performance GPU computing using the CUDA toolkit. This effort includes developing tools to compile Rust to PTX code and create libraries that facilitate the use of existing CUDA libraries in Rust. The project addresses historical issues with Rust's compatibility with CUDA and seeks to advance Rust's role in GPU computing. Contributions are welcome as the project is actively being developed and rebooted.

See all Performance archives