vAttention: Dynamic LLM Memory Without PagedAttention episode artwork

EPISODE · Nov 19, 2025 · 35 MIN

vAttention: Dynamic LLM Memory Without PagedAttention

from The Gist Talk · host kw

paper titled "vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention," introduces a novel memory management approach called vAttention designed to optimize Large Language Model (LLM) serving systems. The paper primarily critiques PagedAttention, the existing standard for dynamic memory allocation, arguing that it introduces performance overheads and complexity by causing the Key-Value (KV) cache to become non-contiguous in virtual memory. vAttention solves this by decoupling virtual and physical memory allocation using CUDA Virtual Memory Management (VMM) APIs, thereby retaining virtual memory contiguity while mitigating physical memory fragmentation. Through evaluations, the authors demonstrate that vAttention is a simpler, more portable, and often more performant alternative, supporting various attention kernels—including FlashAttention-3—out-of-the-box and achieving throughput improvements up to 1.23× over PagedAttention-based systems. The work also details LLM-specific optimizations, such as deferred reclamation and supporting smaller 64KB page groups, to hide VMM latency and reduce fragmentation

Episode metadata supplied by the publisher feed · Published Nov 19, 2025

paper titled "vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention," introduces a novel memory management approach called vAttention designed to optimize Large Language Model (LLM) serving systems. The paper primarily critiques PagedAttention, the existing standard for dynamic memory allocation, arguing that it introduces performance overheads and complexity by causing the Key-Value (KV) cache to become non-contiguous in virtual memory. vAttention solves this by decoupling virtual and physical memory allocation using CUDA Virtual Memory Management (VMM) APIs, thereby retaining virtual memory contiguity while mitigating physical memory fragmentation. Through evaluations, the authors demonstrate that vAttention is a simpler, more portable, and often more performant alternative, supporting various attention kernels—including FlashAttention-3—out-of-the-box and achieving throughput improvements up to 1.23× over PagedAttention-based systems. The work also details LLM-specific optimizations, such as deferred reclamation and supporting smaller 64KB page groups, to hide VMM latency and reduce fragmentation

PodParley-generated summary based on available episode metadata and transcript content.

NOW PLAYING

vAttention: Dynamic LLM Memory Without PagedAttention

0:00 35:01

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

Frequently Asked Questions

How long is this episode of The Gist Talk?

This episode is 35 minutes long.

When was this The Gist Talk episode published?

This episode was published on November 19, 2025.

What is this episode about?

paper titled "vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention," introduces a novel memory management approach called vAttention designed to optimize Large Language Model (LLM) serving systems. The paper primarily...

Can I download this The Gist Talk episode?

Yes, you can download this episode by clicking the download button on the episode player, or subscribe to the podcast in your preferred podcast app for automatic downloads.
URL copied to clipboard!