ABAdarsha Balaji

Adarsha Balaji

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Adarsha Balaji has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evaluation of Test-Time Compute Constraints on Safety and Skill Large Reasoning Models

    by Adarsha Balaji, Le Chen, Rajeev Thakur, Franck Cappello, Sandeep Madireddy


  • Application - Hardware Co-Optimization of Crossbar-Based Neuromorphic Systems

    by Adarsha Balaji, Prasanna Balaprakash

    Spiking Neural Networks (SNNs) executed on neuro-morphic hardware (NmC) have shown great potential to perform a class of learning and inference tasks with low latency and high energy efficiency. However, there are an ever increasing number of design hyperparameters related to the learning algorithms, neuron and synaptic models, neural network architectures and neuromorphic hardware that are required to design the SNN model. These hyperparameters determine the accuracy and en-ergy efficiency of the trained model and often require application, hardware, and software expertise, and are often determined by trial and error, which is a difficult, ad-hoc, and time consuming process. Therefore, there is a need for an automated hyperpa-rameter tuning framework that can search a set of software and hardware hyperparameters to find an SNN model with optimal application accuracy and energy efficiency, when inferred on a NmC. To this end, we propose a framework to co-optimize the accuracy and energy consumption of an SNN model executed on an cross-bar based NmC. The proposed framework integrates three key components: (1) SNNTorch, to train and validate SNN model, (2) SNN-Neurosim, a custom energy estimator for cross-bar based NmC and (3) DeepHyper, a scalable hyperparameter tuning approach to explore the hyperparameter search space of the SNN model and neuromorphic hardware. We evaluate the framework using SNN models trained for scientific applications to (1) parameterize the planetary boundary layer (PBL) using data from the Weather Research Forecast (WRF) model, (2) detect Bragg diffraction peaks for use in X-ray based diffraction microscopy and (3) the reconstruction of high-fidelity amplitude and phase for ptychographic imaging.

  • NeuSB: A Scalable Interconnect Architecture for Spiking Neuromorphic Hardware

    by Adarsha Balaji, Phu Khanh Huynh, Francky Catthoor, Nikil D. Dutt, Jeffrey L. Krichmar, Anup Das

    Neuromorphic systems are typically designed as a tile-based architecture where inter-tile data communication is facilitated using a shared global interconnect. Congestion on this interconnect can increase both interconnect energy, which increases the total energy consumption of the hardware and latency, which impacts the performance e.g., accuracy of the application that is being executed on the hardware. Mesh-based Network-on-Chip (NoC) that is used in most hardware prototypes is not the optimal interconnect solution for neuromorphic systems. This is because of the following two reasons. First, power consumption and average latency of a NoC increases exponentially with the number of tiles in the hardware. Second, a NoC cannot exploit an application’s data communication pattern efficiently. Once designed for a target hardware, the bandwidth on each NoC link stays the same, independent of the volume of data traffic between different tile pairs of the NoC. In other words, a NoC cannot be customized at a finer granularity based on an individual application running on the hardware. We show that these NoC limitations prevent opportunities to further improve energy and latency of a neuromorphic hardware. To address these limitations, we propose Dynamic Segmented Bus (SB) interconnect for neuromorphic systems. Here, a bus lane is partitioned into segments with each segment connecting a few tiles. Connection of tiles to segments and those between segments are bridged using our novel three-way segmentation switches that are programmed using the software before admitting an application to the hardware. We partition an application by analyzing its workload and place partitions intelligently onto segments. This exploits application characteristics to use the segments without any routing collisions while exploiting the latency and energy savings in the design-time mapping phase. At a high-level, our mapping algorithm places tiles that communicate the most on shorter segments utilizing fewer number of switches, thereby reducing network congestion. It can adjust the bandwidth by controlling the number of segments connected to a destination tile. At run time, our controller dynamically executes the predefined routing paths without requiring any additionally routing decisions, unlike a NoC. This allows us to improve both energy and latency. Using parallel segmented busses, our proposed interconnect architecture can support a large number of tiles without significantly increasing the design cost, energy, and latency. Simulation results show that compared to the most widely-used mesh-based NoC design, our interconnect architecture, which we call NeuSB, reduces the switch area by 20x, average interconnect energy by 6.2x, and latency by 23%.


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

    by Adarsha Balaji, Shihao Song, Anup Das, Nikil D. Dutt, Jeffrey Krichmar, Nagarajan Kandasamy, Francky Catthoor

    Neuromorphic hardware with non-volatile memory (NVM) can implement machine learning tasks in an energy-efficient manner. Unfortunately, NVMs such as phase change memory (PCM) require high voltages for correct operation, which are supplied from on-chip charge pumps. If a 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 in the charge pump. Forcefully discharging a stressed charge pump can lower the aging rate of its CMOS devices, but makes the neuromorphic hardware unavailable to perform computations while its charge pumps are being discharged. This negatively impacts performance such as latency and accuracy. 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.


  • RotaSYN: Rotary Traveling Wave Oscillator SYNthesizer

    by Ragh Kuttappa, Adarsha Balaji, Vasil Pano, Baris Taskin, Hamid Mahmoodi

    A physical design methodology is presented to synchronize digital application specific integrated circuit (ASIC) designs by a resonant rotary clock network. One novelty of the proposed RotaSYN flow is that the ASIC products of RotaSYN can operate at previously unattainable (relatively) low-frequency ranges of hundreds of megahertz. The dynamic resonant frequency divider is used to implement the low-frequency operation; and low in comparison to the norm of gigahertz-range of operation for resonant clocking reported in this paper. In SPICE-based simulations, the efficacy of the proposed flow and novel algorithms in RotaSYN is demonstrated using performance metrics of the wirelength, skew, and power on international symposium on physical design-10 clock benchmark circuits. In addition, RotaSYN is compared to three publicly available industrial designs that include the ARM Cortex M0 against equivalent clocks generated with a traditional phase locked loop (PLL) and distributed with an industrial clock tree synthesis tool flow. The RotaSYN methodology is implemented at three different target frequencies of 880 MHz, 500 MHz, and 220 MHz for the industrial designs. SPICE simulations show an average of 29% power savings for the industrial designs overall, solely thanks to 66% power savings on the clock generation and distribution networks, operating at a frequency of 880 MHz on comparison to the PLL-based design with a clock tree synthesized with an industrial EDA tool.