State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper, we first present a new data mapping, called TacitMap, suited for BNNs implemented based on a Computation-In-Memory (CIM) architecture. TacitMap maximizes the use of available parallelism, while CIM architecture eliminates the data movement overhead. We then propose a hardware accelerator based on optical phase change memory (oPCM) called EinsteinBarrier. Ein-steinBarrier incorporates TacitMap and adds an extra dimension for parallelism through wavelength division multiplexing, leading to extra latency reduction. The simulation results show that, compared to the SotA CIM baseline, TacitMap and EinsteinBarrier significantly improve execution time by up to ∼ 154× and ∼ 3113×, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.

Deep Learning (DL) has recently led to remark-able advancements, however, it faces severe computation related challenges. Existing Von-Neumann-based solutions are dealing with issues such as memory bandwidth limitations and energy inefficiency. Computation-In-Memory (CIM) has the potential to address this problem by integrating processing elements directly into the memory architecture, reducing data movement and enhancing the overall efficiency of the system. In this work, we propose CIM architecture using three distinct emerging technologies. Firstly, a CIM architecture utilizing Ferroelectric Field-Effect Transistors (FeFET) is shown and the resulting errors from the analog compute scheme are injected into the emerging algorithm of Hyperdimensional Computing. Subsequently, we explore Vertical Nanowire Field-Effect Transistors (VNWFETs) based CIM within a 3D computing architecture, demonstrating improved energy efficiency and reconfigurability for CIM. Additionally, we improve the accuracy of the Resistive Random Access Memories (RRAM) based CIM architecture using two mapping-based solutions. These three technologies exhibit non-volatile characteristics, and when integrated into the CIM architecture, they yield significant advantages, including enhanced energy efficiency, reliability, and accuracy in computing processes.

Today's computer architectures and semiconductor technologies are facing major challenges making them incapable to deliver the required features (such as computer efficiency) for emerging applications. Alternative architectures are being under investigation in order to continue deliver sustainable benefits for the foreseeable future society at affordable cost. These architectures are not only changing the traditional computing paradigm (e.g., in terms of programming models, compilers, circuit design), but also setting up new challenges and opportunities concerning the test and reliability. This tutorial targets the challenges and opportunities of using approximate computing for achieving low cost fault tolerance mechanisms.

This work presents an approximate triple modular redundancy (ATMR) method based on successive approximation and loop perforation. Successive approximation consists of loop-based algorithms that improve their output quality on every iteration execution. By manipulating the number of executed iterations, a designer can balance output quality and execution time performance. The proposed technique exploits this approximate computing paradigm combined with data size and precision reduction to minimize traditional triple modular redundancy (TMR) costs while maintaining a good fault masking capability. A number of ATMR configurations are presented and tested under laser fault injection to evaluate their fault masking capability. The benchmarks are implemented as embedded software in the ARM Cortex A9 processor of a Xilinx Zynq-7000 series board. The proposed technique was able to improve the execution time overhead with the usage of approximations, with an acceptable fault masking rate.