NKNirmal Kumbhare

Nirmal Kumbhare

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Nirmal Kumbhare has a total of 48 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    CEDR: A Compiler-integrated, Extensible DSSoC Runtime

    by Joshua Mack, Sahil Hassan, Nirmal Kumbhare, Miguel Castro Gonzalez, Ali Akoglu

    In this work, we present a C ompiler-integrated, E xtensible D omain Specific System on Chip R untime (CEDR) ecosystem to facilitate research toward addressing the challenges of architecture, system software, and application development with distinct plug-and-play integration points in a unified compile time and runtime workflow. We demonstrate the utility of CEDR on the Xilinx Zynq MPSoC-ZCU102 for evaluating performance of pre-silicon hardware in the trade space of SoC configuration, scheduling policy and workload complexity based on dynamically arriving workload scenarios composed of real-life signal processing applications scaling to thousands of application instances with Fast Fourier Transform and matrix multiply accelerators. We provide insights into the tradeoffs present in this design space through a number of distinct case studies. CEDR is portable and has been deployed and validated on Odroid-XU3, X86, and Nvidia Jetson Xavier-based SoC platforms. Taken together, CEDR is a capable environment for enabling research in exploring the boundaries of productive application development, resource management heuristic development, and hardware configuration analysis for heterogeneous architectures.


  • RANC: Reconfigurable Architecture for Neuromorphic Computing

    by Joshua Mack, Ruben Purdy, Kris Rockowitz, Michael Inouye, Edward Richter, Spencer Valancius, Nirmal Kumbhare, Sahil Hassan, Kaitlin Fair, John Mixter, et al.

    Neuromorphic architectures have been introduced as platforms for energy-efficient spiking neural network execution. The massive parallelism offered by these architectures has also triggered interest from nonmachine learning application domains. In order to lift the barriers to entry for hardware designers and application developers, we present RANC: a reconfigurable architecture for neuromorphic computing, an opensource highly flexible ecosystem that enables rapid experimentation with neuromorphic architectures in both software via C++ simulation and hardware via FPGA emulation. We present the utility of the RANC ecosystem by showing its ability to recreate behavior of IBM’s TrueNorth and validate with a direct comparison to IBM’s Compass simulation environment and published literature. RANC allows optimizing architectures based on application insights as well as prototyping future neuromorphic architectures that can support new classes of applications entirely. We demonstrate the highly parameterized and configurable nature of RANC by studying the impact of architectural changes on improving application mapping efficiency with quantitative analysis based on Alveo U250 FPGA. We present post routing resource usage and throughput analysis across implementations of synthetic aperture radar classification and vector matrix multiplication applications, and demonstrate a neuromorphic architecture that scales to emulating 259K distinct neurons and 73.3M distinct synapses.


  • Dynamic power management for value-oriented schedulers in power-constrained HPC system

    by Nirmal Kumbhare, Ali Akoglu, Aniruddha Marathe, Salim Hariri, Ghaleb Abdulla


  • DS3: A System-Level Domain-Specific System-on-Chip Simulation Framework

    by Samet Arda, Anish Nk, Ahmet Alper Goksoy, Joshua Mack, Nirmal Kumbhare, Anderson Luiz Sartor, Ali Akoglu, Radu Marculescu, Umit Y. Ogras

    Heterogeneous systems-on-chip (SoCs) are highly favorable computing platforms due to their superior performance and energy efficiency potential compared to homogeneous architectures. They can be further tailored to a specific domain of applications by incorporating processing elements (PEs) that accelerate frequently used kernels in these applications. However, this potential is contingent upon optimizing the SoC for the target domain and utilizing its resources effectively at run-time. To this end, system-level design - including scheduling, power-thermal management algorithms and design space exploration studies - plays a crucial role. This paper presents a system-level domain-specific SoC simulation (DS3) framework to address this need. DS3 enables both design space exploration and dynamic resource management for power-performance optimization of domain applications. We showcase DS3 using six real-world applications from wireless communications and radar processing domain. DS3, as well as the reference applications, is shared as open-source software to stimulate research in this area.


  • Open Access

    User-Space Emulation Framework for Domain-Specific SoC Design

    by Joshua Mack, Nirmal Kumbhare, Anish Nk, Umit Y. Ogras, Ali Akoglu

    In this work, we propose a portable, Linux-based emulation framework to provide an ecosystem for hardware-software co-design of Domain-specific SoCs (DSSoCs) and enable their rapid evaluation during the pre-silicon design phase. This framework holistically targets three key challenges of DSSoC design: accelerator integration, resource management, and application development. We address these challenges via a flexible and lightweight user-space runtime environment that enables easy integration of new accelerators, scheduling heuristics, and user applications, and we illustrate the utility of each through various case studies. With signal processing (WiFi and RADAR) as the target domain, we use our framework to evaluate the performance of various dynamic workloads on hypothetical DSSoC hardware configurations composed of mixtures of CPU cores and FFT accelerators using a Zynq UltraScale+TM MPSoC. We show the portability of this framework by conducting a similar study on an Odroid platform composed of big.LITTLE ARM clusters. Finally, we introduce a prototype compilation toolchain that enables automatic mapping of unlabeled C code to DSSoC platforms. Taken together, this environment offers a unique ecosystem to rapidly perform functional verification and obtain performance and utilization estimates that help accelerate convergence towards a final DSSoC design.

  • Open Access

    FPGA Based Emulation Environment for Neuromorphic Architectures

    by Spencer Valancius, Edward Richter, Ruben Purdy, Kris Rockowitz, Michael Inouye, Joshua Mack, Nirmal Kumbhare, Kaitlin Fair, John Mixter, Ali Akoglu

    Neuromorphic architectures such as IBM's TrueNorth and Intel's Loihi have been introduced as platforms for energy efficient spiking neural network execution. However, there is no framework that allows for rapidly experimenting with neuromorphic architectures and studying the trade space on hardware performance and network accuracy. Fundamentally, this creates a barrier to entry for hardware designers looking to explore neuromorphic architectures. In this paper we present an open-source FPGA based emulation environment for neuromorphic computing research. We prototype IBM's TrueNorth architecture as a reference design and discuss FPGA specific design decisions made when implementing and integrating it's core components. We conduct resource utilization analysis and realize a streaming-enabled TrueNorth architecture on the Zynq UltraScale+ MPSoC. We then perform functional verification by implementing networks for MNIST dataset and vector matrix multiplication (VMM) in our emulation environment and present an accuracy-based comparison based on the same networks generated using IBM's Compass simulation environment. We demonstrate the utility of our emulation environment for hardware designers and application engineers by altering the neuron behavior for VMM mapping, which is, to the best of our knowledge, not feasible with any other tool including IBM's Compass environment. The proposed parameterized and configurable emulation platform serves as a basis for expanding its features to support emerging architectures, studying hypothetical neuromorphic architectures, or rapidly converging to hardware configuration through incremental changes based on bottlenecks as they become apparent during application mapping process.

  • Open Access

    DS3: A System-Level Domain-Specific System-on-Chip Simulation Framework

    by Samet E. Arda, Anish Nk, A. Alper Goksoy, Nirmal Kumbhare, Joshua Mack, Anderson L. Sartor, Ali Akoglu, Radu Marculescu, Umit Y. Ogras

    Heterogeneous systems-on-chip (SoCs) are highly favorable computing platforms due to their superior performance and energy efficiency potential compared to homogeneous architectures. They can be further tailored to a specific domain of applications by incorporating processing elements (PEs) that accelerate frequently used kernels in these applications. However, this potential is contingent upon optimizing the SoC for the target domain and utilizing its resources effectively at runtime. To this end, system-level design - including scheduling, power-thermal management algorithms and design space exploration studies - plays a crucial role. This paper presents a system-level domain-specific SoC simulation (DS3) framework to address this need. DS3 enables both design space exploration and dynamic resource management for power-performance optimization of domain applications. We showcase DS3 using six real-world applications from wireless communications and radar processing domain. DS3, as well as the reference applications, is shared as open-source software to stimulate research in this area.

  • A Value-Oriented Job Scheduling Approach for Power-Constrained and Oversubscribed HPC Systems

    by Nirmal Kumbhare, Aniruddha Marathe, Ali Akoglu, Howard Jay Siegel, Ghaleb Abdulla, Salim Hariri

    In this article, we investigate limitations in the traditional value-based algorithms for a power-constrained HPC system and evaluate their impact on HPC productivity. We expose the trade-off between allocating system-wide power budget uniformly and greedily under different system-wide power constraints in an oversubscribed system. We experimentally demonstrate that, under the tightest power constraint, the mean productivity of the greedy allocation is 38 percent higher than the uniform allocation whereas, under the intermediate power constraint, the uniform allocation has a mean productivity of 6 percent higher than the greedy allocation. We then propose a new algorithm that adapts its behavior to deliver the combined benefits of the two allocation strategies. We design a methodology with online retraining capability to create application-specific power-execution time models for a class of HPC applications. These models are used in predicting the execution time of an application on the available resources at the time of making scheduling decisions in the power-aware algorithms. We evaluate the proposed algorithm using emulation and simulation environments, and show that our adaptive strategy results in improving HPC resource utilization while delivering a mean productivity that is almost the same as the best performing algorithm across various system-wide power constraints.


  • Utility-based resource management in an oversubscribed energy-constrained heterogeneous environment executing parallel applications

    by Dylan Machovec, Bhavesh Khemka, Nirmal Kumbhare, Sudeep Pasricha, Anthony A. Maciejewski, Howard Jay Siegel, Ali Akoglu, Gregory A. Koenig, Salim Hariri, Cihan Tunc, et al.


  • Value of Service Based Task Scheduling for Cloud Computing Systems

    by Cihan Tunc, Nirmal Kumbhare, Ali Akoglu, Salim Hariri, Dylan Machovec, Howard Jay Siegel

    Task scheduling for large scale computing systems is a challenging problem. From the users' perspective, the main concern is the performance of the submitted tasks, whereas, for the cloud service providers, reducing cost while providing the required service is critical. Therefore, there is a need for task scheduling mechanisms that balance users' performance requirements while being energy efficient. We present a time dependent Value of Service (VoS) metric that takes into consideration the arrival time of a task while evaluating the value of completing a task within its deadline and its energy consumption within a constraint. We consider the variation in value for completing a task at different times such that the value of energy reduction can change significantly between peak and non-peak periods. We use completion time and energy based value functions with soft and hard thresholds. Thus, we define the VoS for a given workload to be the sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines (VMs), where each task will be assigned a resource configuration characterized by the number of the homogeneous cores and amount of memory. Using VoS, we design, evaluate, and compare our task scheduling methods to show a significant improvement in energy consumption when considering time-of-use dependent scheduling algorithms. The experimental results are run on an IBM blade server using KVM. The experimental results show using time dependent VoS, 50% improvement in performance value, 40% improvement in energy value, and up to 91% improvement in VoS over a heuristic that only considers only the hard threshold of the value functions.