• TRM (USD/COP)COP 3,461.92 +1.37% +COP 46.82Colombia's official exchange rate (Financial Superintendence), valid Nov 21 to Nov 23
  • EuroCOP 3,804.64 +1.37% +COP 51.45Reference: TRM crossed with the European Central Bank rate of Nov 20; not the market quote
  • Mexican pesoCOP 181.56 +1.37% +COP 2.45Reference: TRM crossed with the European Central Bank rate of Nov 20; not the market quote
← Back to the community
Technology

Broadcom (NASDAQ: AVGO) Tomahawk 6 & Arista vs. Cisco (NASDAQ: CSCO) Silicon One G300: Who Wins 102.4T AI Ethernet?

Both Broadcom's (NASDAQ: AVGO) Tomahawk 6—deployed on Arista Networks (NYSE: ANET) EOS switches alongside sub-250ns Tomahawk Ultra scale-up chips—and Cisco's (NASDAQ: CSCO) Silicon One G300 deliver 102.4 Tbps per chip via 200G SerDes under Ultra Ethernet Consortium (UEC) and IEEE 802.3dj standards for 100,000+ XPU AI clusters.

Per SEC filings and Open Compute Project (OCP) specifications, each 102.4 Tbps ASIC drives 64 ports of 1.6T or 128 ports of 800G Ethernet to challenge Nvidia's (NASDAQ: NVDA) Spectrum-X. Broadcom's Tomahawk Ultra (BCM78920, 51.2 Tbps) enables lossless rack-level scale-up across 256 XPUs using Link Layer Retry, while Cisco's G300 uses a fully shared packet buffer for 28% faster AI job completion.

For hyperscale cloud architects, the multi-billion-dollar fabric decision comes down to Broadcom and Arista's ultra-low cut-through latency and cognitive routing versus Cisco Silicon One's hardware-based congestion telemetry and zero-packet-drop buffering. Which 102.4 Tbps Ethernet silicon will capture more AI data center spine deployments by 2027?

Which 102.4 Tbps switch silicon will lead AI data center Ethernet?

0 votes

Comments

Be respectful. Comments reported by several people are hidden until a moderator reviews them.

You do not need an account to read, only to comment. Sign in with a code sent to your email or with your Google account. We only show the name you choose, never your email.