latency | byteiota

Tag: latency

Split-screen comparison of TCP head-of-line blocking versus Homa SRPT scheduling for AI cluster networking

News

Homa Cuts TCP Latency 13x in AI Clusters — Here’s Why

Stanford's Homa protocol cuts TCP tail latency 13x for AI clusters. The GPU idle cost ...

October 5, 2026

Data visualization comparing Cerebras WSE-3 at 750 tokens per second versus Nvidia H100 at 70 tokens per second for GPT-5.6 Sol inference

News

Cerebras + GPT-5.6 Sol: 750 tok/s Changes Your Agent Latency

Cerebras runs OpenAI GPT-5.6 Sol at 750 tokens per second on WSE-3 — 10x faster ...

July 7, 2026

Diagram showing NUMA non-uniform memory access architecture with dual CPU sockets and local vs remote memory latency paths

Cloud & DevOps

NUMA Explained: Why Memory Distance Slows Your Servers

NUMA creates a 4-5x memory latency gap on multi-socket servers. Learn how first-touch allocation and ...

June 29, 2026