# How we shipped 15 Tbps for OpenAI in 90 days (Session 2 of 3)

DevFeed: [How we shipped 15 Tbps for OpenAI in 90 days (Session 2 of 3)](<https://devfeed.tech/articles/how-we-shipped-15-tbps-for-openai-in-90-days-session-2-of-3-34018.md>)

Original publisher: [Read original article](<https://sridharrajarao.com/blog/openai-15-tbps-session-2/>)

Author: Sridhar Rajarao

Published: 2026-09-12T00:00:00Z

Content type: article

Language: en

Sources: [Sridhar Rajarao](<https://devfeed.tech/sources/sridhar-rajarao.md>)

Topics: [OpenAI](<https://devfeed.tech/topics/openai.md>), [Network](<https://devfeed.tech/topics/network.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [Server](<https://devfeed.tech/topics/server.md>), [Cache](<https://devfeed.tech/topics/cache.md>), [API](<https://devfeed.tech/topics/api.md>), [Oracle Database](<https://devfeed.tech/topics/oracle-database.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [build](<https://devfeed.tech/tags/build.md>), [cache](<https://devfeed.tech/tags/cache.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [database](<https://devfeed.tech/tags/database.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [execution](<https://devfeed.tech/tags/execution.md>), [gateway](<https://devfeed.tech/tags/gateway.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [network](<https://devfeed.tech/tags/network.md>), [object](<https://devfeed.tech/tags/object.md>), [openai](<https://devfeed.tech/tags/openai.md>), [performance](<https://devfeed.tech/tags/performance.md>), [platform-engineering](<https://devfeed.tech/tags/platform-engineering.md>), [server](<https://devfeed.tech/tags/server.md>), [servers](<https://devfeed.tech/tags/servers.md>), [sre](<https://devfeed.tech/tags/sre.md>), [storage](<https://devfeed.tech/tags/storage.md>), [testing](<https://devfeed.tech/tags/testing.md>), [warp](<https://devfeed.tech/tags/warp.md>)

## AI overview

The second session describes turning an architecture for OpenAI's 15 Tbps system into a delivery plan. It covers coordinated capacity planning across network, gateway, server, storage, and database teams; caching object names through the Inventory API; delivery tracking; and performance validation. Early WARP testing found packet drops caused by an unsuitable MTU of 1500, which was changed to 9100.

## Source excerpt

Architecture was only the first week. Session 2 is about the build: capacity, execution discipline, and the first signs that performance would be the real test.