distributed-computing

Who Needs DRAM? We Have Fiber

Weekly Paper Notes — one of the top picks from the 2026-07-11 CS paper digest. Area: Distributed Computing / Systems. Authors: Hannah Atmer, Thiemo Voigt, Yuan Yao, Stefanos Kaxiras (Uppsala University) arXiv: 2607.08407 · PDF TL;DR DRAM is the choke point of the current generative-AI buildout. HBM3e stacks are backordered, contract prices are up, and hyperscalers are absorbing a large fraction of global DRAM output just to fan the same model weights out to ever more accelerators....

Hash-linked blocks sealed with proof-of-work

Bitcoin: A Peer-to-Peer Electronic Cash System

Weekly Paper Notes — Seminal Paper of the Week for the 2026-06-27 CS paper digest. Area: Distributed Computing. Author: Satoshi Nakamoto Published: October 31, 2008 (cryptography mailing list) Canonical URL: bitcoin.org/bitcoin.pdf This week’s digest had a fresh paper on transaction-fair ordering on DAGs without weak edges (Tilikum, 2606.27250) — a piece of work whose entire problem statement (DeFi, BEV, reordering attacks) only exists because of one nine-page document published seventeen-and-a-half years ago to a small cryptography mailing list....

ARGUS architecture: three-channel collection (CPU stack, framework semantics, kernel) feeding a unified pipeline into Grafana and Perfetto

ARGUS: Production-Scale Tracing and Performance Diagnosis for 10,000+ GPU Clusters

Weekly Paper Notes — one of the top picks from the 2026-06-20 CS paper digest. Area: Distributed Computing. Authors: Jiasheng Zhou, Longbin Zeng, Clavis Chen, Ruiming Lu et al. arXiv: 2606.20374 · PDF TL;DR ARGUS is a tracing and performance-diagnosis system designed for always-on operation on production LLM training clusters with more than 10,000 GPUs. The central insight is that no single profiler can be cheap, deep, and continuous all at once — so ARGUS decomposes observation along the training call hierarchy into three independent collection channels: CPU call stacks, framework semantics, and GPU kernel execution....

Clipping Makes Distributed and Federated Asynchronous SGD Robust to Stragglers

Weekly Paper Notes — one of the top picks from the 2026-06-13 CS paper digest. Area: Distributed Computing. Authors: Samuel Erickson, Mikael Johansson (KTH) arXiv: 2606.13287 · PDF TL;DR In asynchronous SGD (ASGD), workers compute gradients on possibly stale parameters and push updates without waiting for slow peers. That’s how you keep all the GPUs busy, but it’s also how slow workers (“stragglers”) inject large delays into the update stream, which classical analyses say should slow convergence in proportion to the maximum delay across the workers....