arXiv subjects

Nagarajan Kandasamy

Publications and source records attributed to Nagarajan Kandasamy.

27 records · Page 2Linked to original sources

arXiv2020

Improving Dependability of Neuromorphic Computing With Non-Volatile Memory

As process technology continues to scale aggressively, circuit aging in a neuromorphic hardware due to negative bias temperature instability (NBTI) and time-dependent dielectric breakdown (TDDB) is becoming a critical reliability issue and is expected to proliferate when using non-volatile memory (NVM) for synaptic storage. This is because an NVM requires high voltage and current to access its synaptic weight, which further accelerates the circuit aging in a neuromorphic hardware. Current methods for qualifying reliability are overly conservative, since they estimate circuit aging considering worst-case operating conditions and unnecessarily constrain performance. This paper proposes RENEU, a reliability-oriented approach to map machine learning applications to neuromorphic hardware, with the aim of improving system-wide reliability without compromising key performance metrics such as execution time of these applications on the hardware. Fundamental to RENEU is a novel formulation of the aging of CMOS-based circuits in a neuromorphic hardware considering different failure mechanisms. Using this formulation, RENEU develops a system-wide reliability model which can be used inside a design-space exploration framework involving the mapping of neurons and synapses to the hardware. To this end, RENEU uses an instance of Particle Swarm Optimization (PSO) to generate mappings that are Pareto-optimal in terms of performance and reliability. We evaluate RENEU using different machine learning applications on a state-of-the-art neuromorphic hardware with NVM synapses. Our results demonstrate an average 38\% reduction in circuit aging, leading to an average 18% improvement in the lifetime of the hardware compared to current practices. RENEU only introduces a marginal performance overhead of 5% compared to a performance-oriented state-of-the-art.

cs.NE↗

arXiv2020

Compiling Spiking Neural Networks to Neuromorphic Hardware

Machine learning applications that are implemented with spike-based computation model, e.g., Spiking Neural Network (SNN), have a great potential to lower the energy consumption when they are executed on a neuromorphic hardware. However, compiling and mapping an SNN to the hardware is challenging, especially when compute and storage resources of the hardware (viz. crossbar) need to be shared among the neurons and synapses of the SNN. We propose an approach to analyze and compile SNNs on a resource-constrained neuromorphic hardware, providing guarantee on key performance metrics such as execution time and throughput. Our approach makes the following three key contributions. First, we propose a greedy technique to partition an SNN into clusters of neurons and synapses such that each cluster can fit on to the resources of a crossbar. Second, we exploit the rich semantics and expressiveness of Synchronous Dataflow Graphs (SDFGs) to represent a clustered SNN and analyze its performance using Max-Plus Algebra, considering the available compute and storage capacities, buffer sizes, and communication bandwidth. Third, we propose a self-timed execution-based fast technique to compile and admit SNN-based applications to a neuromorphic hardware at run-time, adapting dynamically to the available resources on the hardware. We evaluate our approach with standard SNN-based applications and demonstrate a significant performance improvement compared to current practices.

cs.DC↗

arXiv2020

Improving Phase Change Memory Performance with Data Content Aware Access

A prominent characteristic of write operation in Phase-Change Memory (PCM) is that its latency and energy are sensitive to the data to be written as well as the content that is overwritten. We observe that overwriting unknown memory content can incur significantly higher latency and energy compared to overwriting known all-zeros or all-ones content. This is because all-zeros or all-ones content is overwritten by programming the PCM cells only in one direction, i.e., using either SET or RESET operations, not both. In this paper, we propose data content aware PCM writes (DATACON), a new mechanism that reduces the latency and energy of PCM writes by redirecting these requests to overwrite memory locations containing all-zeros or all-ones. DATACON operates in three steps. First, it estimates how much a PCM write access would benefit from overwriting known content (e.g., all-zeros, or all-ones) by comprehensively considering the number of set bits in the data to be written, and the energy-latency trade-offs for SET and RESET operations in PCM. Second, it translates the write address to a physical address within memory that contains the best type of content to overwrite, and records this translation in a table for future accesses. We exploit data access locality in workloads to minimize the address translation overhead. Third, it re-initializes unused memory locations with known all-zeros or all-ones content in a manner that does not interfere with regular read and write accesses. DATACON overwrites unknown content only when it is absolutely necessary to do so. We evaluate DATACON with workloads from state-of-the-art machine learning applications, SPEC CPU2017, and NAS Parallel Benchmarks. Results demonstrate that DATACON significantly improves system performance and memory system energy consumption compared to the best of performance-oriented state-of-the-art techniques.

cs.AR↗

arXiv2020

Exploiting Inter- and Intra-Memory Asymmetries for Data Mapping in Hybrid Tiered-Memories

Modern computing systems are embracing hybrid memory comprising of DRAM and non-volatile memory (NVM) to combine the best properties of both memory technologies, achieving low latency, high reliability, and high density. A prominent characteristic of DRAM-NVM hybrid memory is that it has NVM access latency much higher than DRAM access latency. We call this inter-memory asymmetry. We observe that parasitic components on a long bitline are a major source of high latency in both DRAM and NVM, and a significant factor contributing to high-voltage operations in NVM, which impact their reliability. We propose an architectural change, where each long bitline in DRAM and NVM is split into two segments by an isolation transistor. One segment can be accessed with lower latency and operating voltage than the other. By introducing tiers, we enable non-uniform accesses within each memory type (which we call intra-memory asymmetry), leading to performance and reliability trade-offs in DRAM-NVM hybrid memory. We extend existing NVM-DRAM OS in three ways. First, we exploit both inter- and intra-memory asymmetries to allocate and migrate memory pages between the tiers in DRAM and NVM. Second, we improve the OS's page allocation decisions by predicting the access intensity of a newly-referenced memory page in a program and placing it to a matching tier during its initial allocation. This minimizes page migrations during program execution, lowering the performance overhead. Third, we propose a solution to migrate pages between the tiers of the same memory without transferring data over the memory channel, minimizing channel occupancy and improving performance. Our overall approach, which we call MNEME, to enable and exploit asymmetries in DRAM-NVM hybrid tiered memory improves both performance and reliability for both single-core and multi-programmed workloads.

cs.AR↗

arXiv2019

A Framework to Explore Workload-Specific Performance and Lifetime Trade-offs in Neuromorphic Computing

Neuromorphic hardware with non-volatile memory (NVM) can implement machine learning workload in an energy-efficient manner. Unfortunately, certain NVMs such as phase change memory (PCM) require high voltages for correct operation. These voltages are supplied from an on-chip charge pump. If the charge pump is activated too frequently, its internal CMOS devices do not recover from stress, accelerating their aging and leading to negative bias temperature instability (NBTI) generated defects. Forcefully discharging the stressed charge pump can lower the aging rate of its CMOS devices, but makes the neuromorphic hardware unavailable to perform computations while its charge pump is being discharged. This negatively impacts performance such as latency and accuracy of the machine learning workload being executed. In this paper, we propose a novel framework to exploit workload-specific performance and lifetime trade-offs in neuromorphic computing. Our framework first extracts the precise times at which a charge pump in the hardware is activated to support neural computations within a workload. This timing information is then used with a characterized NBTI reliability model to estimate the charge pump's aging during the workload execution. We use our framework to evaluate workload-specific performance and reliability impacts of using 1) different SNN mapping strategies and 2) different charge pump discharge strategies. We show that our framework can be used by system designers to explore performance and reliability trade-offs early in the design of neuromorphic hardware such that appropriate reliability-oriented design margins can be set.

cs.ET↗

arXiv2019

Enabling and Exploiting Partition-Level Parallelism (PALP) in Phase Change Memories

Phase-change memory (PCM) devices have multiple banks to serve memory requests in parallel. Unfortunately, if two requests go to the same bank, they have to be served one after another, leading to lower system performance. We observe that a modern PCM bank is implemented as a collection of partitions that operate mostly independently while sharing a few global peripheral structures, which include the sense amplifiers (to read) and the write drivers (to write). Based on this observation, we propose PALP, a new mechanism that enables partition-level parallelism within each PCM bank, and exploits such parallelism by using the memory controller's access scheduling decisions. PALP consists of three new contributions. First, we introduce new PCM commands to enable parallelism in a bank's partitions in order to resolve the read-write bank conflicts, with minimal changes needed to PCM logic and its interface. Second, we propose simple circuit modifications that introduce a new operating mode for the write drivers, in addition to their default mode of serving write requests. When configured in this new mode, the write drivers can resolve the read-read bank conflicts, working jointly with the sense amplifiers. Finally, we propose a new access scheduling mechanism in PCM that improves performance by prioritizing those requests that exploit partition-level parallelism over other requests, including the long outstanding ones. While doing so, the memory controller also guarantees starvation-freedom and the PCM's running-average-power-limit (RAPL). We evaluate PALP with workloads from the MiBench and SPEC CPU2017 Benchmark suites. Our results show that PALP reduces average PCM access latency by 23%, and improves average system performance by 28% compared to the state-of-the-art approaches.

cs.AR↗

arXiv2016

FPGA Implementation of a Scalable and Run-Time Adaptable Multi-Standard Packet Detector

This paper describes a step by step approach for implementing a scalable and run-time adaptable multi-standard packet detector for orthogonal frequency divisional multiplexing (OFDM) based communication standards. The paper briefly describes considerations and design choices in making a modular system block with generic control supporting rapid prototyping and implementation of preamble-based packet detectors. The results were generated through implementation on a Xilinx Virtex-6 FPGA with a MicroBlaze processor instantiated to provide run-time control and adaptability.

cs.NI↗

arXiv2016

A New Approach to Dimensionality Reduction for Anomaly Detection in Data Traffic

The monitoring and management of high-volume feature-rich traffic in large networks offers significant challenges in storage, transmission and computational costs. The predominant approach to reducing these costs is based on performing a linear mapping of the data to a low-dimensional subspace such that a certain large percentage of the variance in the data is preserved in the low-dimensional representation. This variance-based subspace approach to dimensionality reduction forces a fixed choice of the number of dimensions, is not responsive to real-time shifts in observed traffic patterns, and is vulnerable to normal traffic spoofing. Based on theoretical insights proved in this paper, we propose a new distance-based approach to dimensionality reduction motivated by the fact that the real-time structural differences between the covariance matrices of the observed and the normal traffic is more relevant to anomaly detection than the structure of the training data alone. Our approach, called the distance-based subspace method, allows a different number of reduced dimensions in different time windows and arrives at only the number of dimensions necessary for effective anomaly detection. We present centralized and distributed versions of our algorithm and, using simulation on real traffic traces, demonstrate the qualitative and quantitative advantages of the distance-based subspace approach.

cs.LG↗

arXiv2016

Experimental Evaluation of a Reconfigurable Antenna System for Blind Interference Alignment

In recent years, several experimental studies have come out to validate the theoretical findings of interference alignment (IA), but only a handful of studies have focused on blind interference alignment. Unlike IA and other interference mitigation techniques, blind IA does not require channel state information at the transmitter (CSIT). The key insight is that the transmitter uses the knowledge of channel coherence intervals and receivers utilize reconfigurable antennas to create channel fluctuations exploited by the transmitter. In this work, we present a novel experimental evaluation of a reconfigurable antenna system for achieving blind IA. We present a blind IA technique based on reconfigurable antennas for a 2-user multipleinput single-output (MISO) broadcast channel implemented on a software defined radio platform where each of the receivers is equipped with a reconfigurable antenna. We further compare this blind IA implementation with traditional TDMA scheme for benchmarking purposes. We show that the achievable rates for blind IA can be realized in practice using measured channels under practical channel conditions. Additionally, the average error vector magnitude and bit error rate (BER) performances are evaluated.

cs.IT↗