# Blueprint for Supercharging LLM Inference With "PagedAttention over RDMA"

**Published:** April 18, 2025

![Blueprint for Supercharging LLM Inference With "PagedAttention over RDMA"](https://cdn.sanity.io/images/ult5g8gw/production/355f68b0903c8cff111cfcf84d397a5e6a861523-960x540.jpg)

**Watch:** https://fast.wistia.net/embed/iframe/wrjhzilfq9

PagedAttention over RDMA" (PAoR) revolutionizes large language model serving by addressing key-value (KV) cache challenges with RDMA networking

## Commentary

Learn how "PagedAttention over RDMA" (PAoR) revolutionizes large language model serving by addressing key-value (KV) cache challenges with RDMA networking and NeuralMesh. This session showcases seamless integration with vLLM and TensorRT-LLM, enabling faster inference with reduced latency and increased throughput across multi-node environments.

## Related Videos

- [Win Back Wasted GPU Compute With KV Cache Efficiency](/video/win-back-wasted-gpu-compute-with-kv-cache-efficiency)
- [The New Math of Inference](/video/the-new-math-of-inference)
- [Why Inference Economics Are Reshaping Neoclouds](/video/why-inference-economics-are-reshaping-neoclouds)
