Fastrack: Fast IO for Secure ML using GPU TEEs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yongqin, Rajat, Rachit, Lee, Jonghyun, Tang, Tingting, Annavaram, Murali
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916447480971264
author Wang, Yongqin
Rajat, Rachit
Lee, Jonghyun
Tang, Tingting
Annavaram, Murali
author_facet Wang, Yongqin
Rajat, Rachit
Lee, Jonghyun
Tang, Tingting
Annavaram, Murali
contents As cloud-based ML expands, ensuring data security during training and inference is critical. GPU-based Trusted Execution Environments (TEEs) offer secure, high-performance solutions, with CPU TEEs managing data movement and GPU TEEs handling authentication and computation. However, CPU-to-GPU communication overheads significantly hinder performance, as data must be encrypted, authenticated, decrypted, and verified, increasing costs by 12.69 to 33.53 times. This results in GPU TEE inference becoming 54.12% to 903.9% slower and training 10% to 455% slower than non-TEE systems, undermining GPU TEE advantages in latency-sensitive applications. This paper analyzes Nvidia H100 TEE protocols and identifies three key overheads: 1) redundant CPU re-encryption, 2) limited authentication parallelism, and 3) unnecessary operation serialization. We propose Fastrack, optimizing with 1) direct GPU TEE communication, 2) parallelized authentication, and 3) overlapping decryption with PCI-e transmission. These optimizations cut communication costs and reduce inference/training runtime by up to 84.6%, with minimal overhead compared to non-TEE systems.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15240
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fastrack: Fast IO for Secure ML using GPU TEEs
Wang, Yongqin
Rajat, Rachit
Lee, Jonghyun
Tang, Tingting
Annavaram, Murali
Cryptography and Security
Hardware Architecture
As cloud-based ML expands, ensuring data security during training and inference is critical. GPU-based Trusted Execution Environments (TEEs) offer secure, high-performance solutions, with CPU TEEs managing data movement and GPU TEEs handling authentication and computation. However, CPU-to-GPU communication overheads significantly hinder performance, as data must be encrypted, authenticated, decrypted, and verified, increasing costs by 12.69 to 33.53 times. This results in GPU TEE inference becoming 54.12% to 903.9% slower and training 10% to 455% slower than non-TEE systems, undermining GPU TEE advantages in latency-sensitive applications. This paper analyzes Nvidia H100 TEE protocols and identifies three key overheads: 1) redundant CPU re-encryption, 2) limited authentication parallelism, and 3) unnecessary operation serialization. We propose Fastrack, optimizing with 1) direct GPU TEE communication, 2) parallelized authentication, and 3) overlapping decryption with PCI-e transmission. These optimizations cut communication costs and reduce inference/training runtime by up to 84.6%, with minimal overhead compared to non-TEE systems.
title Fastrack: Fast IO for Secure ML using GPU TEEs
topic Cryptography and Security
Hardware Architecture
url https://arxiv.org/abs/2410.15240