Secure IVSHMEM: End-to-End Shared-Memory Protocol with Hypervisor-CA Handshake and In-Kernel Access Control

📅 2025-05-25
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
IVSHMEM lacks built-in security mechanisms, rendering it vulnerable to eavesdropping and tampering—thus unsuitable for high-assurance environments. This paper proposes an end-to-end trusted shared memory protocol that achieves mutual authentication and fine-grained access control for zero-copy inter-VM communication—a first in the domain. Key contributions include: (1) a lightweight Hypervisor-CA handshake mechanism (<200 ms per session), (2) hardware-assisted channel isolation, (3) kernel-module-level access enforcement, and (4) application-layer abstraction optimizations. Experimental evaluation shows the protocol incurs ≤5% increase in data-plane round-trip latency over baseline IVSHMEM, with negligible bandwidth overhead. It delivers strong security guarantees while preserving near-native performance—making it suitable for safety- and real-time-critical domains such as automotive electronics.

Technology Category

Application Category

📝 Abstract
In-host shared memory (IVSHMEM) enables high-throughput, zero-copy communication between virtual machines, but today's implementations lack any security control, allowing any application to eavesdrop or tamper with the IVSHMEM region. This paper presents Secure IVSHMEM, a protocol that provides end-to-end mutual authentication and fine-grained access enforcement with negligible performance cost. We combine three techniques to ensure security: (1) channel separation and kernel module access control, (2)hypervisor-mediated handshake for end-to-end service authentication, and (3)application-level integration for abstraction and performance mitigation. In microbenchmarks, Secure IVSHMEM completes its one-time handshake in under 200ms and sustains data-plane round-trip latencies within 5% of the unmodified baseline, with negligible bandwidth overhead. We believe this design is ideally suited for safety and latency-critical in-host domains, such as automotive systems, where both performance and security are paramount.
Problem

Research questions and friction points this paper is trying to address.

Lack of security control in IVSHMEM allowing eavesdropping or tampering
Need for end-to-end mutual authentication in shared-memory protocols
Requirement for fine-grained access enforcement with minimal performance impact
Innovation

Methods, ideas, or system contributions that make the work stand out.

Kernel module enforces fine-grained access control
Hypervisor mediates end-to-end service authentication
Application-level integration reduces performance impact
🔎 Similar Papers
No similar papers found.
H
Hyunwoo Kim
Intel Corporation
Jaeseong Lee
Jaeseong Lee
KAIST
Deep LearningComputer VisionComputer Graphics
S
Sunpyo Hong
Intel Corporation
C
Changmin Han
Intel Corporation