KYKaige Yan

Kaige Yan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kaige Yan has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • GEREM: Fast and Precise Error Resilience Assessment for GPU Microarchitectures

    by Jingweijia Tan, Xurui Li, An Zhong, Kaige Yan, Xiaohui Wei, Guanpeng Li

    GPUs are widely used hardware acceleration platforms in many areas due to their great computational throughput. In the meanwhile, GPUs are vulnerable to transient hardware faults in the post-Moore era. Analyzing the error resilience of GPUs are critical for both hardware and software. Statistical fault injection approaches are commonly used for error resilience analysis, which are highly accurate but very time consuming. In this work, we propose GEREM, a first framework to speed up fault injection process so as to estimate the error resilience of GPU microarchitectures swiftly and precisely. We find early fault behaviors can be used to accurately predict the final outcomes of program execution. Based on this observation, we categorize the early behaviors of hardware faults into GPU Early Fault Manifestation models (EFMs). For data structures, EFMs are early propagation characteristics of faults, while for pipeline instructions, EFMs are heuristic properties of several instruction contexts. We further observe that EFMs are determined by static microarchitecture states, so we can capture them without actually simulating the program execution process under fault injections. Leveraging these observations, our GEREM framework first profiles the microarchitectural states related for EFMs at one time. It then injects faults into the profiled traces to immediately generate EFMs. For data storage structures, EFMs are directly used to predict final fault outcomes, while for pipeline instructions, machine learning is used for prediction. Evaluation results show GEREM precisely assesses the error resilience of GPU microarchitecture structures with $237\times$ speedup on average comparing with traditional fault injections.


  • Evaluating GPU's Instruction-Level Error Characteristics Under Low Supply Voltages

    by Jingweijia Tan, Jiashuo Wang, Kaige Yan, Xiaohui Wei, Xin Fu

    Supply voltage underscaling has been an effective approach to improve the energy-efficiency of modern high-performance processors, such as GPUs. However, energy efficiency and reliability are two sides of a trade-off. Undervolting will inevitably undermine reliability, since it reduces chip manufacturers’ voltage guardbands that is designed to ensure correct operations under worst-case scenarios. To achieve optimal energy efficiency while maintaining enough reliability, it is necessary to deeply understand the error characteristics caused by undervolting. Unlike previous works which focus mostly on program level, we perform the first comprehensive instruction-level voltage margin and error characteristics evaluation for GPU architectures. We systematically measure the error probability and patterns of GPU instructions during undervolting. Then, we also analyze the impact of locations (SMs, threads, and bits) and operand data values on the error characteristics. Based on our observations, we reduce the voltage to the minimum safe limit for different instructions which achieves 18.37% energy saving, and we further propose an error detection strategy which reduces the performance and energy overhead by 14.8% with negligible 0.01% degradation for error detection rate.


  • Open Access

    Improving the Performance of CNN Accelerator Architecture under the Impact of Process Variations

    by Jingweijia Tan, Weiren Wang, Maodi Ma, Xiaohui Wei, Kaige Yan

    Convolutional neural network (CNN) accelerators are popular specialized platforms for efficient CNN processing. As semiconductor manufacturing technology scales down to nano scale, process variation dramatically affects the chip’s quality. Process variation causes delay variation within the chip due to transistor parameter differences. CNN accelerators adopt a large number of processing elements (PEs) for parallel computing, which are highly susceptible to process variation effects. Fast CNN processing desires consistent performance among PEs; otherwise the processing speed is limited by the slowest PE within the chip. In this work, we first quantitatively model and analyze the impact of process variation on CNN accelerators’ operating frequency. We further analyze the utilization of CNN accelerators and the characteristics of CNN models. We then leverage the PE underutilization to propose a sub-matrix reformation mechanism and leverage the pixel similarity of images to propose a weight transfer technique. Both techniques are able to tolerate the low-frequency PEs and achieve performance improvement at chip level. Furthermore, a novel resilience-aware mapping technique that exploits the diversity in the importance of weights is also proposed to improve the performance. Evaluation results show that our techniques are able to achieve significant processing speed improvement with negligible accuracy loss.


  • Saca-AVF: A Quantitative Approach to Analyze the Architectural Vulnerability Factors of CNN Accelerators

    by Jingweijia Tan, Liqi Ping, Qixiang Wang, Kaige Yan

    Convolutional neural network (CNN) accelerators are widely used in artificial intelligence applications, such as image recognitions, due to their superior computing performance. However, as manufacturing technology scales down, the shrinking of chip size and the increasing of integration density make CNN accelerators vulnerable to soft errors, which are key factors affecting the reliability of integrated circuits. For emerging CNN applications where reliability is critical (such as self-driving cars), visible faults in applications' outputs caused by soft errors may lead to catastrophic consequences. Therefore, it is important to consider reliability into CNN accelerators' architecture design. Recent modeling and fault injection approaches analyze the reliability features of CNN models, but none evaluate the soft error reliability of CNN accelerators from the architecture level. The Architectural Vulnerability Factor (AVF) indicates the probability that a soft error results in faulty computation outcomes. The Architecturally Correct Execution (ACE) method analyzes AVF of structures by identifying ACE bits and calculating their residency time. Based on the traditional ACE-based AVF analysis model, we propose a quantitative approach saca-AVF to analyze the AVF of systolic array based CNN accelerators that perform homogeneous matrix arithmetic operations instead of various kinds of instructions. We explore the reliability features of CNN accelerators under different design choices at different levels leveraging saca-AVF. We also observe several reliability characteristics of the CNN accelerator architecture. We believe our proposed saca-AVF and the observations we made are able to guide designers to discover the reliability hotspots of CNN accelerators and build highly reliable CNN accelerators.


  • Architectural Design Model Guided On-Demand Power Management of Energy-Efficient GPGPU for SLAM

    by Kaige Yan, Zhujun Ma, Caiwei Li, Xin Fu, Jingweijia Tan

    Simultaneously localization and mapping (SLAM) is a core component in many embedded domains, e.g., robots, augmented and virtual reality. Due to SLAM’s high demand on computation resources, general-purpose graphic processing units (GPGPUs) are often used as its processing engine. Meanwhile, embedded systems usually have strict power constraint. Thus, how to deliver required performance for SLAM, yet still meet the power limit, is a great challenge faced by GPGPU designer. In this work, we discover the general principles of designing energy-efficient GPGPU for SLAM as “many SMs, enough SPs and registers, small caches”, by analyzing the implication of individual design parameters on both performance and power. Then, we conduct large-scale design space exploration and fit the Pareto frontier with a two-term exponential model. Further, we construct gradient boosting decision tree (GBDT)-based design models to predict the performance and power given the design parameters. The evaluation shows that our GBDT-based models can achieve [ ]3% mean average percentage error, which significantly outperform other machine learning models. With these models, a kernel’s requirement on hardware resources can be well understood. Based on such knowledge, we introduce design model guided power management strategies, including power gating and dynamic frequency and voltage scaling (DFVS). Overall, by combining these two power management strategies, we can improve the energy delay product by 36%.