arXiv subjects

Yiqi Chen

Publications and source records attributed to Yiqi Chen.

27 records · Page 2Linked to original sources

arXiv2024

Theseus: Exploring Efficient Wafer-Scale Chip Design for Large Language Models

The emergence of the large language model~(LLM) poses an exponential growth of demand for computation throughput, memory capacity, and communication bandwidth. Such a demand growth has significantly surpassed the improvement of corresponding chip designs. With the advancement of fabrication and integration technologies, designers have been developing Wafer-Scale Chips~(WSCs) to scale up and exploit the limits of computation density, memory capacity, and communication bandwidth at the level of a single chip. Existing solutions have demonstrated the significant advantages of WSCs over traditional designs, showing potential to effectively support LLM workloads. Despite the benefits, exploring the early-stage design space of WSCs for LLMs is a crucial yet challenging task due to the enormous and complicated design space, time-consuming evaluation methods, and inefficient exploration strategies. To address these challenges, we propose Theseus, an efficient WSC design space exploration framework for LLMs. We construct the design space of WSCs with various constraints considering the unique characteristics of WSCs. We propose efficient evaluation methodologies for large-scale NoC-based WSCs and introduce multi-fidelity Bayesian optimization to efficiently explore the design space. Evaluation results demonstrate the efficiency of Theseus that the searched Pareto optimal results outperform GPU cluster and existing WSC designs by up to 62.8\%/73.7\% in performance and 38.6\%/42.4\% in power consumption for LLM training, while improving up to 23.2$\times$ and 15.7$\times$ for the performance and power of inference tasks. Furthermore, we conduct case studies to address the design tradeoffs in WSCs and provide insights to facilitate WSC designs for LLMs.

cs.AR↗

arXiv2024

Distribution-Preserving Integrated Sensing and Communication with Secure Reconstruction

Distribution-preserving integrated sensing and communication with secure reconstruction is investigated in this paper. In addition to the distortion constraint, we impose another constraint on the distance between the reconstructed sequence distribution and the original state distribution to force the system to preserve the statistical property of the channel states. An inner bound of the distribution-preserving capacity-distortion region is provided with some capacity region results under special cases. A numerical example demonstrates the tradeoff between the communication rate, reconstruction distortion and distribution preservation. Furthermore, we consider the case that the reconstructed sequence should be kept secret from an eavesdropper who also observes the channel output. An inner bound of the tradeoff region and a capacity-achieving special case are presented.

cs.IT↗

arXiv2024

NeoMem: Hardware/Software Co-Design for CXL-Native Memory Tiering

The Compute Express Link (CXL) interconnect makes it feasible to integrate diverse types of memory into servers via its byte-addressable SerDes links. Considering the various access latency, harnessing the full potential of CXL-based heterogeneous memory systems requires efficient memory tiering. However, prior work can hardly make a fundamental progress owing to low-resolution and high-overhead memory access profiling techniques. To address this critical challenge, we propose a novel memory tiering solution called NeoMem, which features a hardware/software co-design. NeoMem offloads memory profiling functions to CXL device-side controllers, integrating a dedicated hardware unit called NeoProf. NeoProf readily monitors memory accesses and provides the OS with crucial page hotness statistics and other useful system state information. On the OS kernel side, we design a revamped memory-tiering strategy, enabling accurate and timely hot page promotion based on NeoProf statistics. We implement NeoMem on a real FPGA-based CXL memory platform and Linux kernel v6.3. Comprehensive evaluations demonstrate that NeoMem achieves 32% to 67% geomean speedup over several existing memory tiering solutions.

cs.AR↗

arXiv2024

The Dawn of AI-Native EDA: Opportunities and Challenges of Large Circuit Models

Within the Electronic Design Automation (EDA) domain, AI-driven solutions have emerged as formidable tools, yet they typically augment rather than redefine existing methodologies. These solutions often repurpose deep learning models from other domains, such as vision, text, and graph analytics, applying them to circuit design without tailoring to the unique complexities of electronic circuits. Such an AI4EDA approach falls short of achieving a holistic design synthesis and understanding, overlooking the intricate interplay of electrical, logical, and physical facets of circuit data. This paper argues for a paradigm shift from AI4EDA towards AI-native EDA, integrating AI at the core of the design process. Pivotal to this vision is the development of a multimodal circuit representation learning technique, poised to provide a comprehensive understanding by harmonizing and extracting insights from varied data sources, such as functional specifications, RTL designs, circuit netlists, and physical layouts. We champion the creation of large circuit models (LCMs) that are inherently multimodal, crafted to decode and express the rich semantics and structures of circuit data, thus fostering more resilient, efficient, and inventive design methodologies. Embracing this AI-native philosophy, we foresee a trajectory that transcends the current innovation plateau in EDA, igniting a profound shift-left in electronic design methodology. The envisioned advancements herald not just an evolution of existing EDA tools but a revolution, giving rise to novel instruments of design tools that promise to radically enhance design productivity and inaugurate a new epoch where the optimization of circuit performance, power, and area (PPA) is achieved not incrementally, but through leaps that redefine the benchmarks of electronic systems' capabilities.

cs.AR↗

arXiv2023

On General Capacity-Distortion Formulas of Integrated Sensing and Communication

The integrated sensing and communication (ISAC) problem with general state and channel distributions is investigated. General formulas of the capacity-distortion tradeoff for the ISAC problem under maximal and average distortion constraints are provided. The results cover some existing communication models such as the general point-to-point channel and Gel'fand-Pinsker channel. More details including memoryless states and channels, mixed states and channels, and rate-limited CSI at one side are considered. Numerical results focus on writing-on-dirty paper model and channel with ergodic/non-ergodic fading.

cs.IT↗

arXiv2023

On Strong Secrecy for Multiple Access Channels with States and Causal CSI

Strong secrecy communication over a discrete memoryless state-dependent multiple access channel (SD-MAC) with an external eavesdropper is investigated. The channel is governed by discrete memoryless and i.i.d. channel states and the channel state information (CSI) is revealed to the encoders in a causal manner. Inner and outer bounds are provided. To establish the inner bound, we investigate coding schemes incorporating wiretap coding and secret key agreement between the sender and the legitimate receiver. Two kinds of block Markov coding schemes are proposed. The first one is a new coding scheme using backward decoding and Wyner-Ziv coding and the secret key is constructed from a lossy description of the CSI. The other one is an extended version of the existing coding scheme for point-to-point wiretap channels with causal CSI. A numerical example shows that the achievable region given by the first coding scheme can be strictly larger than the second one. However, these two schemes do not outperform each other in general and there exists some numerical examples that in different channel models each coding scheme achieves some rate pairs that cannot be achieved by another scheme. Our established inner bound reduces to some best-known results in the literature as special cases. We further investigate some capacity-achieving cases for state-dependent multiple access wiretap channels (SD-MAWCs) with degraded message sets. It turns out that the two coding schemes are both optimal in these cases.

cs.IT↗

arXiv2023

On Strong Secrecy for Multiple Access Channel with States and Causal CSI

cs.IT↗

arXiv2021

PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization

Quantization is one of the most effective methods to compress neural networks, which has achieved great success on convolutional neural networks (CNNs). Recently, vision transformers have demonstrated great potential in computer vision. However, previous post-training quantization methods performed not well on vision transformer, resulting in more than 1% accuracy drop even in 8-bit quantization. Therefore, we analyze the problems of quantization on vision transformers. We observe the distributions of activation values after softmax and GELU functions are quite different from the Gaussian distribution. We also observe that common quantization metrics, such as MSE and cosine distance, are inaccurate to determine the optimal scaling factor. In this paper, we propose the twin uniform quantization method to reduce the quantization error on these activation values. And we propose to use a Hessian guided metric to evaluate different scaling factors, which improves the accuracy of calibration at a small cost. To enable the fast quantization of vision transformers, we develop an efficient framework, PTQ4ViT. Experiments show the quantized vision transformers achieve near-lossless prediction accuracy (less than 0.5% drop at 8-bit quantization) on the ImageNet classification task.

cs.CV↗

arXiv2021

PTQ-SL: Exploring the Sub-layerwise Post-training Quantization

Network quantization is a powerful technique to compress convolutional neural networks. The quantization granularity determines how to share the scaling factors in weights, which affects the performance of network quantization. Most existing approaches share the scaling factors layerwisely or channelwisely for quantization of convolutional layers. Channelwise quantization and layerwise quantization have been widely used in various applications. However, other quantization granularities are rarely explored. In this paper, we will explore the sub-layerwise granularity that shares the scaling factor across multiple input and output channels. We propose an efficient post-training quantization method in sub-layerwise granularity (PTQ-SL). Then we systematically experiment on various granularities and observe that the prediction accuracy of the quantized neural network has a strong correlation with the granularity. Moreover, we find that adjusting the position of the channels can improve the performance of sub-layerwise quantization. Therefore, we propose a method to reorder the channels for sub-layerwise quantization. The experiments demonstrate that the sub-layerwise quantization with appropriate channel reordering can outperform the channelwise quantization.

cs.CV↗