skip to main content

Explore Research Products in the NSF-PAR

Title: RCoal: Mitigating GPU Timing Attack via Subwarp-Based Randomized Coalescing Techniques
Graphics processing units (GPUs) are becoming default accelerators in many domains such as high-performance computing (HPC), deep learning, and virtual/augmented reality. Recently, GPUs have also shown significant speedups for a variety of security-sensitive applications such as encryptions. These speedups have largely benefited from the high memory bandwidth and compute throughput of GPUs. One of the key features to optimize the memory bandwidth consumption in GPUs is intra-warp memory access coalescing, which merges memory requests originating from different threads of a single warp into as few cache lines as possible. However, this coalescing feature is also shown to make the GPUs prone to the correlation timing attacks as it exposes the relationship between the execution time and the number of coalesced accesses. Consequently, an attacker is able to correctly reveal an AES private key via repeatedly gathering encrypted data and execution time on a GPU. In this work, we propose a series of defense mechanisms to alleviate such timing attacks by carefully trading off performance for improved security. Specifically, we propose to randomize the coalescing logic such that the attacker finds it hard to guess the correct number of coalesced accesses generated. To this end, we propose to randomize: a) the granularity (called as subwarp) at which warp threads are grouped together for coalescing, and b) the threads selected by each subwarp for coalescing. Such randomization techniques result in three mechanisms: fixed-sized subwarp (FSS), random-sized subwarp (RSS), and random-threaded subwarp (RTS). We find that the combination of these security mechanisms offers 24- to 961-times improvement in the security against the correlation timing attacks with 5 to 28% performance degradation. Online copy: http://adwaitjog.github.io/docs/pdf/rcoal-hpca18.pdf  more » « less
Award ID(s):
1717532
NSF-PAR ID:
10059888
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
2018 IEEE International Symposium on High Performance Computer Architecture (HPCA)
Page Range / eLocation ID:
156 to 167
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ( , 2022 IEEE International Symposium on Secure and Private Execution Environment Design)
    Abstract—Recent work has demonstrated the security risk associated with micro-architecture side-channels. The cache timing side-channel is a particularly popular target due to its availability and high leakage bandwidth. Existing proposals for defending cache side-channel attacks either degrade cache performance and/or limit cache sharing, hence, should only be invoked when the system is under attack. A lightweight monitoring mechanism that detects malicious micro-architecture manipulation in realistic environments is essential for the judicious deployment of these defense mechanisms. In this paper, we propose PREDATOR, a cache side-channel attack detector that identifies cache events caused by an attacker. To detect side-channel attacks in noisy environments, we take advantage of the observation that, unlike non-specific noises, an active attacker alters victim’s micro-architectural states on security critical accesses and thus causes the victim extra cache events on those accesses. PREDATOR uses precise performance counters to collect detailed victim’s access information and analyzes location-based deviations. PREDATOR is capable of detecting five different attacks with high accuracy and limited performance overhead in complex noisy execution environments. PREDATOR remains effective even when the attacker slows the attack rate by 256 times. Furthermore, PREDATOR is able to accurately report details about the attack such as the instruction that accesses the attacked data. In the case of GnuPG RSA [20], PREDATOR can pinpoint the square/multiply operations in the Modulo-Reduce algorithm; and in the case of OpenSSL AES [45], it can identify the accesses to the Te-Table. 
    more » « less
  2. ; ; ; ; ; ( , Network and Distributed System Security Symposium)
    Program obfuscation is a popular cryptographic construct with a wide range of uses such as IP theft prevention. Although cryptographic solutions for program obfuscation impose impractically high overheads, a recent breakthrough leveraging trusted hardware has shown promise. However, the existing solution is based on special-purpose trusted hardware, restricting its use-cases to a limited few. In this paper, we first study if such obfuscation is feasible based on commodity trusted hardware, Intel SGX, and we observe that certain important security considerations are not afforded by commodity hardware. In particular, we found that existing obfuscation/obliviousness schemes are insecure if directly applied to Intel SGX primarily due to side-channel limitations. To this end, we present OBFUSCURO, the first system providing program obfuscation using commodity trusted hardware, Intel SGX. The key idea is to leverage ORAM operations to perform secure code execution and data access. Initially, OBFUSCURO transforms the regular program layout into a side-channel secure and ORAM-compatible layout. Then, OBFUSCURO ensures that its ORAM controller performs data oblivious accesses in order to protect itself from all memory-based side-channels. Furthermore, OBFUSCURO ensures that the program is secure from timing attacks by ensuring that the program always runs for a pre-configured time interval. Along the way, OBFUSCURO also introduces a systematic optimization such as register-based ORAM stash. We provide a thorough security analysis of OBFUSCURO along with empirical attack evaluations showing that OBFUSCURO can protect the SGX program execution from being leaked by access pattern-based and timing-based channels. We also provide a detailed performance benchmark results in order to show the practical aspects of OBFUSCURO. 
    more » « less
  3. ; ; ; ; ; ; ; ( , Proceedings of the Twenty-Third International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '18))
    Graphics Processing Units (GPUs) exploit large amounts of thread-level parallelism to provide high instruction throughput and to efficiently hide long-latency stalls. The resulting high throughput, along with continued programmability improvements, have made GPUs an essential computational resource in many domains. Applications from different domains can have vastly different compute and memory demands on the GPU. In a large-scale computing environment, to efficiently accommodate such wide-ranging demands without leaving GPU resources underutilized, multiple applications can share a single GPU, akin to how multiple applications execute concurrently on a CPU. Multi-application concurrency requires several support mechanisms in both hardware and software. One such key mechanism is virtual memory, which manages and protects the address space of each application. However, modern GPUs lack the extensive support for multi-application concurrency available in CPUs, and as a result suffer from high performance overheads when shared by multiple applications, as we demonstrate. We perform a detailed analysis of which multi-application concurrency support limitations hurt GPU performance the most. We find that the poor performance is largely a result of the virtual memory mechanisms employed in modern GPUs. In particular, poor address translation performance is a key obstacle to efficient GPU sharing. State-of-the-art address translation mechanisms, which were designed for single-application execution, experience significant inter-application interference when multiple applications spatially share the GPU. This contention leads to frequent misses in the shared translation lookaside buffer (TLB), where a single miss can induce long-latency stalls for hundreds of threads. As a result, the GPU often cannot schedule enough threads to successfully hide the stalls, which diminishes system throughput and becomes a first-order performance concern. Based on our analysis, we propose MASK, a new GPU framework that provides low-overhead virtual memory support for the concurrent execution of multiple applications. MASK consists of three novel address-translation-aware cache and memory management mechanisms that work together to largely reduce the overhead of address translation: (1) a token-based technique to reduce TLB contention, (2) a bypassing mechanism to improve the effectiveness of cached address translations, and (3) an application-aware memory scheduling scheme to reduce the interference between address translation and data requests. Our evaluations show that MASK restores much of the throughput lost to TLB contention. Relative to a state-of-the-art GPU TLB, MASK improves system throughput by 57.8%, improves IPC throughput by 43.4%, and reduces application-level unfairness by 22.4%. MASK's system throughput is within 23.2% of an ideal GPU system with no address translation overhead. 
    more » « less
  4. ; ; ; ( , ACM Transactions on Architecture and Code Optimization)
    null (Ed.)
    Contemporary GPUs support multiple kernels to run concurrently on the same streaming multiprocessors (SMs). Recent studies have demonstrated that such concurrent kernel execution (CKE) improves both resource utilization and computational throughput. Most of the prior works focus on partitioning the GPU resources at the cooperative thread array (CTA) level or the warp scheduler level to improve CKE. However, significant performance slowdown and unfairness are observed when latency-sensitive kernels co-run with bandwidth-intensive ones. The reason is that bandwidth over-subscription from bandwidth-intensive kernels leads to much aggravated memory access latency, which is highly detrimental to latency-sensitive kernels. Even among bandwidth-intensive kernels, more intensive kernels may unfairly consume much higher bandwidth than less-intensive ones. In this article, we first make a case that such problems cannot be sufficiently solved by managing CTA combinations alone and reveal the fundamental reasons. Then, we propose a coordinated approach for CTA combination and bandwidth partitioning. Our approach dynamically detects co-running kernels as latency sensitive or bandwidth intensive. As both the DRAM bandwidth and L2-to-L1 Network-on-Chip (NoC) bandwidth can be the critical resource, our approach partitions both bandwidth resources coordinately along with selecting proper CTA combinations. The key objective is to allocate more CTA resources for latency-sensitive kernels and more NoC/DRAM bandwidth resources to NoC-/DRAM-intensive kernels. We achieve it using a variation of dominant resource fairness (DRF). Compared with two state-of-the-art CKE optimization schemes, SMK [52] and WS [55], our approach improves the average harmonic speedup by 78% and 39%, respectively. Even compared to the best possible CTA combinations, which are obtained from an exhaustive search among all possible CTA combinations, our approach improves the harmonic speedup by up to 51% and 11% on average. 
    more » « less
  5. ; ; ( , ACM Transactions on Cyber-Physical Systems)
    null (Ed.)
    Defense mechanisms against network-level attacks are commonly based on the use of cryptographic techniques, such as lengthy message authentication codes (MAC) that provide data integrity guarantees. However, such mechanisms require significant resources (both computational and network bandwidth), which prevents their continuous use in resource-constrained cyber-physical systems (CPS). Recently, it was shown how physical properties of controlled systems can be exploited to relax these stringent requirements for systems where sensor measurements and actuator commands are transmitted over a potentially compromised network; specifically, that merely intermittent use of data authentication (i.e., at occasional time points during system execution), can still provide strong Quality-of-Control (QoC) guarantees even in the presence of false-data injection attacks, such as Man-in-the-Middle (MitM) attacks. Consequently, in this work, we focus on integrating security into existing resource-constrained CPS, in order to protect against MitM attacks on a system where a set of control tasks communicates over a real-time network with system sensors and actuators. We introduce a design-time methodology that incorporates requirements for QoC in the presence of attacks into end-to-end timing constraints for real-time control transactions, which include data acquisition and authentication, real-time network messages, and control tasks. This allows us to formulate a mixed integer linear programming-based method for direct synthesis of schedulable tasks and message parameters (i.e., deadlines and offsets) that do not violate timing requirements for the already deployed controllers, while adding a sufficient level of protection against network-based attacks; specifically, the synthesis method also provides suitable intermittent authentication policies that ensure the desired QoC levels under attack. To additionally reduce the security-related bandwidth overhead, we propose the use of cumulative message authentication at time instances when the integrity of messages from subsets of sensors should be ensured. Furthermore, we introduce a method for the opportunistic use of the remaining resources to further improve the overall QoC guarantees while ensuring system (i.e., task and message) schedulability. Finally, we demonstrate applicability and scalability of our methodology on synthetic automotive systems as well as a real-world automotive case-study. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email