Lightweight JavaScript library that predicts user intent based on mouse movements and keyboard navigation. Perfect for smart prefetching. Also works for mobile!

Coroutine-Oriented Main-Memory Database Engine (VLDB 2021)

🖱 Preload Angular lazy-loaded routes on mouse over

A speculative mechanism to accelerate long-latency off-chip load requests by removing on-chip cache access latency from their critical path, as described by MICRO 2022 paper by Bera et al. (https://arxiv.org/pdf/2209.00188.pdf)

Development repository for Fetch Directed Instruction Prefetching (FDP) in gem5

Hermeto is a CLI tool that prefetches your project dependencies to aid in making your container build process hermetic.

This project focuses on simulating a multi-tier storage system🔺, with an emphasis on data management📂🔄 through the implementation of various placement policies 🔺⬆️⬇️ . The goal is to analyze the performance and efficiency of different data placement and eviction strategies📊⚡🚀 .

Task-based framework for automatic synchronization and memory prefetching.

When caching is not enough. Pre-fetch everything you need to RAM and refresh it on a regular basis.

Instruction Pointer Classifier and Dynamic Degree Stream based Hardware Cache Prefetching

A JavaScript client-side predictive prefetch solution for Webpack and SPAs. An alternative to Guess.js.

A high-performance cache and memory hierarchy simulator built with modern C++17. Features configurable cache levels, advanced prefetching, MESI protocol, and detailed statistics. Ideal for computer architecture education, research, and performance analysis.

A data prefetching storage data plane for accelerating DL training performance.

⚡️Faster subsequent page-loads by prefetching in-viewport links during idle time for React, port of https://getquick.link

The ultimate browser performance library for lightning-fast web experiences

Implementation of a navigation-aware technique for personalized prefetching of network requests of Android apps

VectorFFT is a vectorized, pure C FFT library optimized for x86 processors (AVX-512, AVX2, SSE2) with zero external dependencies. It implements mixed-radix algorithms for common sizes and Bluestein's method for arbitrary lengths, with OpenMP multi-threading for large transforms. Designed for both digital signal processing and financial applications

foresightjs implementation for Angular

Task-based DBMS prototype using MxTasking and query compilation.

A reader for Substack