面向可靠的基于TCAM存内类脑超维计算的硬件/软件协同设计
HW/SW Co-design for Reliable TCAM-based In-memory Brain-inspired Hyperdimensional Computing
AI总结:
本文研究不精确存内硬件对HDC推理的影响,建模SRAM与FeFET中电压相关错误,并通过软硬件协同设计实现最高6倍节能及面积、功耗收益。
摘要:
类脑超维计算(HDC)正持续获得显著关注。由于其能够从少量数据中学习、实现轻量级且具备错误韧性,它是传统机器学习方法的一种有前景的替代方案。然而,与传统机器学习算法类似,HDC也极度以数据为中心。存内计算正迅速兴起,它通过消除计算单元与存储单元之间的数据移动来克服冯·诺依曼瓶颈。在本工作中,我们研究并建模了不精确的存内计算硬件对HDC推理精度的影响。我们的建模基于14nm FinFET工艺,并使用Intel测量数据进行了完全校准。我们首次精确建模了基于SRAM和基于FeFET的存内计算中与电压相关的错误概率。得益于HDC对错误的韧性,底层硬件的复杂度可以降低,从而带来最高达6倍的大幅节能。针对SRAM的实验结果表明,由工艺变化引起的错误概率最高可达39%。尽管错误概率如此之高,推理精度仅受到轻微影响。这为探索新的权衡关系打开了大门。我们还证明,对错误的韧性依赖于具体应用。此外,当底层存内硬件使用新兴的非易失性FeFET器件而非成熟的基于CMOS的SRAM实现时,我们研究了HDC对错误的鲁棒性。我们证明,尽管错误概率更大,推理精度确实仍保持较高水平,同时可获得显著的面积和功耗节省。总而言之,对于传统CMOS工艺和即将到来的新兴技术而言,硬件/软件协同设计都是实现高效且可靠的存内超维计算的关键。
英文摘要:
Brain-inspired hyperdimensional computing (HDC) is continuously gaining remarkable attention. It is a promising alternative to traditional machine-learning approaches due to its ability to learn from little data, lightweight implementation, and resiliency against errors. However, HDC is overwhelmingly data-centric similar to traditional machine-learning algorithms. In-memory computing is rapidly emerging to overcome the von Neumann bottleneck by eliminating data movements between compute and storage units. In this work, we investigate and model the impact of imprecise in-memory computing hardware on the inference accuracy of HDC. Our modeling is based on 14nm FinFET technology fully calibrated with Intel measurement data. We accurately model, for the first time, the voltage-dependent error probability in SRAM-based and FeFET-based in-memory computing. Thanks to HDC's resiliency against errors, the complexity of the underlying hardware can be reduced, providing large energy savings of up to 6x. Experimental results for SRAM reveal that variability-induced errors have a probability of up to 39 percent. Despite such a high error probability, the inference accuracy is only marginally impacted. This opens doors to explore new tradeoffs. We also demonstrate that the resiliency against errors is application-dependent. In addition, we investigate the robustness of HDC against errors when the underlying in-memory hardware is realized using emerging non-volatile FeFET devices instead of mature CMOS-based SRAMs. We demonstrate that inference accuracy does remain high despite the larger error probability, while large area and power savings can be obtained. All in all, HW/SW co-design is the key for efficient yet reliable in-memory hyperdimensional computing for both conventional CMOS technology and upcoming emerging technologies.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。