Skip to content
EpisodeWednesday, May 6, 2026 · 14 min
$episode№026·date2026-05-06·duration14 min·turns71

Astera Labs Redefines AI Scale-Up Fabric as KV Cache Becomes the Infrastructure Bottleneck

Read the briefing
Astera Labs Redefines AI Scale-Up Fabric as KV Cache Becomes the Infrastructure Bottleneck
12 sources · quality 4.5/5

Astera Labs launches the three-hundred-twenty-lane Scorpio X-Series memory-semantic fabric switch as two simultaneous research papers reveal that KV cache management has overtaken compute as the defining bottleneck for agentic AI infrastructure. NVIDIA's Vera Rubin codesign data shows agents consume fifteen times more tokens than chat, making today's capacity models systematically wrong for 2027 workloads.

0:00/0:00loading
Transcript
71 turns · ~11 min read
HOST A

Welcome to Amaze Networks for Wednesday, May sixth.

HOST B

Can I jump straight in? Because on Monday we covered the Broadcom Thor Ultra and this whole debate about whether Ethernet wins the AI fabric war. Today there's a direct answer to that question, and it comes from an unexpected direction.

HOST A

Yeah, Astera Labs dropped the Scorpio X-Series this week and I've been thinking about it since the announcement landed. Walk us through it.

Subscribe

New episodes, every weekday.

Amaze Networks drops at 4 AM CT, Monday through Friday. Spotify and Apple Podcasts submissions in progress.

RSS FeedSpotify · soonApple · soonEmail — read instead