The wide usage of hardware intellectual property cores from untrusted vendors has raised security concerns for system designers. Existing solutions for functionality testing and verification do not ...usually consider the presence of malicious logic in hardware. Formal methods provide powerful solutions for detecting malicious behaviors in hardware. However, they suffer from scalability issues and cannot be easily used for large-scale computing systems. To alleviate the scalability challenge, we propose a new integrated formal verification framework to evaluate the trust of system-on-chip (SoC) constructed from untrusted third-party hardware resources. This framework combines an automated model checker with an interactive theorem prover to reduce the time for proving the system-level security properties of SoCs. Another factor contributing to the scalability issue is the effort required for manual conversion of the hardware design from register transfer level (RTL) code to a domain-specific language prior to verification. Consequently, we develop an automatic code converter for translating VHSIC hardware description language (VHDL) to Formal-HDL, which is a domain specific language for representing hardware designs in the language of Coq. To demonstrate the effectiveness of our integrated verification framework and automated code conversion tool, we evaluate a vulnerable program executed on a bare metal LEON3 SPARC V8 processor and prove system security with considerable reduction in verification effort.
Transformer is a deep learning language model widely used for natural language processing (NLP) services in datacenters. Among transformer models, Generative Pre-trained Transformer (GPT) has ...achieved remarkable performance in text generation, or natural language generation (NLG), which needs the processing of a large input context in the summarization stage, followed by the generation stage that produces a single word at a time. The conventional platforms such as GPU are specialized for the parallel processing of large inputs in the summarization stage, but their performance significantly degrades in the generation stage due to its sequential characteristic. Therefore, an efficient hardware platform is required to address the high latency caused by the sequential characteristic of text generation.
In this paper, we present DFX, a multi-FPGA acceleration appliance that executes GPT-2 model inference end-to-end with low latency and high throughput in both summarization and generation stages. DFX uses model parallelism and optimized dataflow that is model-and-hardware-aware for fast simultaneous workload execution among devices. Its compute cores operate on custom instructions and provide GPT-2 operations end-to-end. We implement the proposed hardware architecture on four Xilinx Alveo U280 FPGAs and utilize all of the channels of the high bandwidth memory (HBM) and the maximum number of compute resources for high hardware efficiency. DFX achieves 5.58× speedup and 3.99× energy efficiency over four NVIDIA V100 GPUs on the modern GPT-2 model. DFX is also 8.21× more cost-effective than the GPU appliance, suggesting that it is a promising solution for text generation workloads in cloud datacenters.
Bouquet of instruction pointers Pakalapati, Samuel; Panda, Biswabandan
2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA),
05/2020
Conference Proceeding
Hardware prefetching is one of the common off-chip DRAM latency hiding techniques. Though hardware prefetchers are ubiquitous in the commercial machines and prefetching techniques are well studied in ...the computer architecture community, the "memory wall" problem still exists after decades of micro-architecture research and is considered to be an essential problem to solve. In this paper, we make a case for breaking the memory wall through data prefetching at the L1 cache.
We propose a bouquet of hardware prefetchers that can handle a variety of access patterns driven by the control flow of an application. We name our proposal Instruction Pointer Classifier based spatial Prefetching (IPCP). We propose IPCP in two flavors: (i) an L1 spatial data prefetcher that classifies instruction pointers at the L1 cache level, and issues prefetch requests based on the classification, and (ii) a multi-level IPCP where the IPCP at the L1 communicates the classification information to the L2 IPCP so that it can kick-start prefetching based on this classification done at the L1. Overall, IPCP is a simple, lightweight, and modular framework for L1 and multi-level spatial prefetching. IPCP at the L1 and L2 incurs a storage overhead of 740 bytes and 155 bytes, respectively.
Our empirical results show that, for memory-intensive single-threaded SPEC CPU 2017 benchmarks, compared to a baseline system with no prefetching, IPCP provides an average performance improvement of 45.1%. For the entire SPEC CPU 2017 suite, it provides an improvement of 22%. In the case of multi-core systems, IPCP provides an improvement of 23.4% (evaluated over more than 1000 mixes). IPCP outperforms the already high-performing state-of-the-art prefetchers like SPP with PPF and Bingo by demanding 30X to 50X less storage.
Homomorphic Encryption (HE) is one of the most promising post-quantum cryptographic schemes that enable privacy-preserving computation on servers. However, noise accumulates as we perform operations ...on HE-encrypted data, restricting the number of possible operations. Fully HE (FHE) removes this restriction by introducing the bootstrapping operation, which refreshes the data; however, FHE schemes are highly memory-bound. Bootstrapping, in particular, requires loading GBs of evaluation keys and plaintexts from off-chip memory, which makes FHE acceleration fundamentally bottlenecked by the off-chip memory bandwidth.
In this paper, we propose ARK, an Accelerator for FHE with Runtime data generation and inter-operation Key reuse. ARK enables practical FHE workloads with a novel algorithm-architecture co-design to accelerate bootstrapping. We first eliminate the off-chip memory bandwidth bottleneck through runtime data generation and inter-operation key reuse. This approach enables ARK to fully exploit on-chip memory by substantially reducing the size of the working set. On top of such algorithmic enhancements, we build ARK microarchitecture that minimizes on-chip data movement through an efficient, alternating data distribution policy based on the data access patterns and a streamlined dataflow organization of the tailored functional units ---including base conversion, number-theoretic transform, and automorphism units. Overall, our co-design effectively handles the heavy computation and data movement overheads of FHE, drastically reducing the cost of HE operations, including bootstrapping.
Fault attacks pose a potent threat to modern cryptographic implementations, particularly those used in physically approachable embedded devices in IoT environments. Information security in such ...resource-constrained devices is ensured using lightweight ciphers, where combinational circuit implementations of SBox are preferable over look-up tables as they are more efficient regarding area, power, and memory requirements. Most existing fault analysis techniques focus on fault injection in memory cells and registers. Recently, a novel fault model and analysis technique, namely Semi-Permanent Stuck-At (SPSA) fault analysis, has been proposed to evaluate the security of ciphers with combinational circuit implementation of Substitution layer elements, SBox. In this work, we propose optimized techniques to recover the key in a minimum number of ciphertexts in such implementations of lightweight ciphers. Based on the proposed techniques, a key recovery attack on the NIST lightweight cryptography (NIST-LWC) standardization process finalist, Elephant AEAD, has been proposed. The proposed key recovery attack is validated on two versions of Elephant cipher. The proposed fault analysis approach recovered the secret key within 85–240 ciphertexts, calculated over 1,000 attack instances. To the best of our knowledge, this is the first work on fault analysis attacks on the Elephant scheme. Furthermore, an optimized combinational circuit implementation of Spongent SBox (SBox used in Elephant cipher) is proposed, having a smaller gate count than the optimized implementation reported in the literature. The proposed fault analysis techniques are validated on primary and optimized versions of Spongent SBox through Verilog simulations. Further, we pinpoint SPSA hotspots in the lightweight GIFT cipher SBox architecture. We observe that GIFT SBox exhibits resilience toward the proposed SPSA fault analysis technique under the single fault adversarial model. However, eight SPSA fault patterns reduce the nonlinearity of the SBox to zero, rendering it vulnerable to linear cryptanalysis. Conclusively, SPSA faults may adversely affect the cryptographic properties of an SBox, thereby leading to trivial key recovery. The GIFT cipher is used as an example to focus on two aspects: (i) its SBox construction is resilient to the proposed SPSA analysis and therefore characterizing such constructions for SPSA resilience and (ii) an SBox even though resilient to the proposed SPSA analysis, may exhibit vulnerabilities toward other classical analysis techniques when subjected to SPSA faults. Our work reports new vulnerabilities in fault analysis in the combinational circuit implementations of cryptographic protocols.
Attention-based neural networks have become pervasive in many AI tasks. Despite their excellent algorithmic performance, the use of the attention mechanism and feedforward network (FFN) demands ...excessive computational and memory resources, which often compromises their hardware performance. Although various sparse variants have been introduced, most approaches only focus on mitigating the quadratic scaling of attention on the algorithm level, without explicitly considering the efficiency of mapping their methods on real hardware designs. Furthermore, most efforts only focus on either the attention mechanism or the FFNs but without jointly optimizing both parts, causing most of the current designs to lack scalability when dealing with different input lengths. This paper systematically considers the sparsity patterns in different variants from a hardware perspective. On the algorithmic level, we propose FABNet, a hardware-friendly variant that adopts a unified butterfly sparsity pattern to approximate both the attention mechanism and the FFNs. On the hardware level, a novel adaptable butterfly accelerator is proposed that can be configured at runtime via dedicated hardware control to accelerate different butterfly layers using a single unified hardware engine. On the Long-Range-Arena dataset, FABNet achieves the same accuracy as the vanilla Transformer while reducing the amount of computation by 10 ~ 66× and the number of parameters 2 ~ 22×. By jointly optimizing the algorithm and hardware, our FPGA-based butterfly accelerator achieves 14.2 ~ 23.2× speedup over state-of-the-art accelerators normalized to the same computational budget. Compared with optimized CPU and GPU designs on Raspberry Pi 4 and Jetson Nano, our system is up to 273.8× and 15.1× faster under the same power budget.
Towards Scaling Blockchain Systems via Sharding Dang, Hung; Dinh, Tien Tuan Anh; Loghin, Dumitrel ...
Proceedings of the 2019 International Conference on Management of Data,
06/2019
Conference Proceeding
Odprti dostop
Existing blockchain systems scale poorly because of their distributed consensus protocols. Current attempts at improving blockchain scalability are limited to cryptocurrency. Scaling blockchain ...systems under general workloads (i.e., non-cryptocurrency applications) remains an open question. This work takes a principled approach to apply sharding to blockchain systems in order to improve their transaction throughput at scale. This is challenging, however, due to the fundamental difference in failure models between databases and blockchain. To achieve our goal, we first enhance the performance of Byzantine consensus protocols, improving individual shards' throughput. Next, we design an efficient shard formation protocol that securely assigns nodes into shards. We rely on trusted hardware, namely Intel SGX, to achieve high performance for both consensus and shard formation protocol. Third, we design a general distributed transaction protocol that ensures safety and liveness even when transaction coordinators are malicious. Finally, we conduct an extensive evaluation of our design both on a local cluster and on Google Cloud Platform. The results show that our consensus and shard formation protocols outperform state-of-the-art solutions at scale. More importantly, our sharded blockchain reaches a high throughput that can handle Visa-level workloads, and is the largest ever reported in a realistic environment.
QUAC-TRNG Olgun, Ataberk; Patel, Minesh; Yağlikçi, A. Giray ...
2021 ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA),
06/2021
Conference Proceeding
Odprti dostop
True random number generators (TRNG) sample random physical processes to create large amounts of random numbers for various use cases, including security-critical cryptographic primitives, scientific ...simulations, machine learning applications, and even recreational entertainment. Unfortunately, not every computing system is equipped with dedicated TRNG hardware, limiting the application space and security guarantees for such systems. To open the application space and enable security guarantees for the overwhelming majority of computing systems that do not necessarily have dedicated TRNG hardware (e.g., processing-in-memory systems), we develop QUAC-TRNG, a new high-throughput TRNG that can be fully implemented in commodity DRAM chips, which are key components in most modern systems.
QUAC-TRNG exploits the new observation that a carefully-engineered sequence of DRAM commands activates four consecutive DRAM rows in rapid succession. This QUadruple ACtivation (QUAC) causes the bitline sense amplifiers to non-deterministically converge to random values when we activate four rows that store conflicting data because the net deviation in bitline voltage fails to meet reliable sensing margins.
We experimentally demonstrate that QUAC reliably generates random values across 136 commodity DDR4 DRAM chips from one major DRAM manufacturer. We describe how to develop an effective TRNG (QUAC-TRNG) based on QUAC. We evaluate the quality of our TRNG using the commonly-used NIST statistical test suite for randomness and find that QUAC-TRNG successfully passes each test. Our experimental evaluations show that QUAC-TRNG reliably generates true random numbers with a throughput of 3.44 Gb/s (per DRAM channel), outperforming the state-of-the-art DRAM-based TRNG by 15.08X and 1.41X for basic and throughput-optimized versions, respectively. We show that QUAC-TRNG utilizes DRAM bandwidth better than the state-of-the-art, achieving up to 2.03X the throughput of a throughput-optimized baseline when scaling bus frequencies to 12 GT/s.
Hardware Trojan attack in the form of malicious modification of a design has emerged as a major security threat. Sidechannel analysis has been investigated as an alternative to conventional logic ...testing to detect the presence of hardware Trojans. However, these techniques suffer from decreased sensitivity toward small Trojans, especially because of the large process variations present in modern nanometer technologies. In this paper, we propose a novel noninvasive, multiple-parameter side-channel analysisbased Trojan detection approach. We use the intrinsic relationship between dynamic current and maximum operating frequency of a circuit to isolate the effect of a Trojan circuit from process noise. We propose a vector generation approach and several design/test techniques to improve the detection sensitivity. Simulation results with two large circuits, a 32-bit integer execution unit (IEU) and a 128-bit advanced encryption standard (AES) cipher, show a detection resolution of 1.12 percent amidst ±20 percent parameter variations. The approach is also validated with experimental results. Finally, the use of a combined side-channel analysis and logic testing approach is shown to provide high overall detection coverage for hardware Trojan circuits of varying types and sizes.
Binary neural networks (BNNs) are promising to deliver accuracy comparable to conventional deep neural networks at a fraction of the cost in terms of memory and energy. In this paper, we introduce ...the XNOR neural engine (XNE), a fully digital configurable hardware accelerator IP for BNNs, integrated within a microcontroller unit (MCU) equipped with an autonomous I/O subsystem and hybrid SRAM/standard cell memory. The XNE is able to fully compute convolutional and dense layers in autonomy or in cooperation with the core in the MCU to realize more complex behaviors. We show post-synthesis results in 65- and 22-nm technology for the XNE IP and post-layout results in 22 nm for the full MCU indicating that this system can drop the energy cost per binary operation to 21.6 fJ per operation at 0.4 V, and at the same time is flexible and performant enough to execute state-of-the-art BNN topologies such as ResNet-34 in less than 2.2 mJ per frame at 8.9 frames/s.