Author

# Janakiraman Viraraghavan

Other affiliations: Indian Institute of Science, IBM, GlobalFoundries

Bio: Janakiraman Viraraghavan is an academic researcher from Indian Institute of Technology Madras. The author has contributed to research in topic(s): Logic gate & Transistor. The author has an hindex of 6, co-authored 16 publication(s) receiving 80 citation(s). Previous affiliations of Janakiraman Viraraghavan include Indian Institute of Science & IBM.

##### Papers

More filters

••

TL;DR: A 1.1 Mb embedded DRAM macro (eDRAM), for next-generation IBM SOI processors, employs 14 nm FinFET logic technology with 0.0174 μm2 deep-trench capacitor cell that enables a high voltage gain of a power-gated inverter at mid-level input voltage.

Abstract: A 1.1 Mb embedded DRAM macro (eDRAM), for next-generation IBM SOI processors, employs 14 nm FinFET logic technology with $\hbox{0.0174}~\mu\hbox{m}^{2}$ deep-trench capacitor cell. A Gated-feedback sense amplifier enables a high voltage gain of a power-gated inverter at mid-level input voltage, while supporting 66 cells per local bit-line. A dynamic-and-gate-thin-oxide word-line driver that tracks standard logic process variation improves the eDRAM array performance with reduced area. The 1.1 $~$ Mb macro composed of 8 $\times$ 2 72 Kb subarrays is organized with a center interface block architecture, allowing 1 ns access latency and 1 ns bank interleaving operation using two banks, each having 2 ns random access cycle. 5 GHz operation has been demonstrated in a system prototype, which includes 6 instances of 1.1 Mb eDRAM macros, integrated with an array-built-in-self-test engine, phase-locked loop (PLL), and word-line high and word-line low voltage generators. The advantage of the 14 nm FinFET array over the 22 nm array was confirmed using direct tester control of the 1.1 Mb eDRAM macros integrated in 16 Mb inline monitor.

16 citations

••

15 Jun 2016

TL;DR: An 80Kb logic Embedded Multi-Time Programmable Memory (MTPM) employs charge trapping and de-trapping behavior in 32nm/22nm High-K transistor, resulting in no added process complexity.

Abstract: An 80Kb logic Embedded Multi-Time Programmable Memory (MTPM) employs charge trapping and de-trapping behavior in 32nm/22nm High-K transistor, resulting in no added process complexity. Multi-step verification with overwrite protection employs block-write and signal margin degradation (∼30%) to satisfy 10 year retention at 105° C.

12 citations

••

TL;DR: The design and implementation of an 80-kb logic-embedded non-volatile multi-time programmable memory (MTPM) with no added process complexity is described and high-temperature stress results show a projected data retention of 10 years at 125 °C.

Abstract: This paper describes the design and implementation of an 80-kb logic-embedded non-volatile multi-time programmable memory (MTPM) with no added process complexity. Charge trap transistors (CTTs) that exploit charge trapping and de-trapping behavior in high-K dielectric of 32-/22-nm Logic FETs are used as storage elements with logic-compatible programming voltages. A high-gain slew-sense amplifier (SA) is used to efficiently detect the threshold voltage difference ( $\Delta V_{\textrm {DIF}}$ ) between the true and complement FETs in the twin cell. Design-assist techniques including multi-step programming with over-write protection and block write algorithm are used to enhance the programming efficiency without causing a dielectric breakdown. High-temperature stress results show a projected data retention of 10 years at 125 °C with a signal loss of <30% that is margined in while programming, by employing a sense margining logic in the SA. Scalability of CTT has been established by the first demonstration of CTT-based MTPM in 14-nm bulk FinFET technology with read cycle time of 40 ns at 0.7-V VDD.

10 citations

••

IBM

^{1}TL;DR: ANNs can model a much higher degree of nonlinearity compared to existing quadratic polynomial models and, hence, can even be used in sub-100-nm technologies to model leakage current that exponentially depends on process parameters.

Abstract: A technique for extracting statistical compact model parameters using artificial neural networks (ANNs) is proposed. ANNs can model a much higher degree of nonlinearity compared to existing quadratic polynomial models and, hence, can even be used in sub-100-nm technologies to model leakage current that exponentially depends on process parameters. Existing techniques cannot be extended to handle such exponential functions. Additionally, ANNs can handle multiple input multiple output relations very effectively. The concept applied to CMOS devices improves the efficiency and accuracy of model extraction. Results from the ANN match the ones obtained from SPICE simulators within 1%.

9 citations

••

04 Jan 2008TL;DR: This work characterize standard cell libraries for statistical leakage analysis based on models for transistor stacks and investigates the use of neural networks for the combined PVT model, for the stacks, which can capture the effect of inter die, intra gate variations, supply voltage and temperature on leakage.

Abstract: With extensive use of dynamic voltage scaling (DVS) there is increasing need for voltage scalable models. Similarly, leakage being very sensitive to temperature motivates the need for a temperature scalable model as well. We characterize standard cell libraries for statistical leakage analysis based on models for transistor stacks. Modeling stacks has the advantage of using a single model across many gates there by reducing the number of models that need to be characterized. Our experiments on 15 different gates show that we needed only 23 models to predict the leakage across 126 input vector combinations. We investigate the use of neural networks for the combined PVT model, for the stacks, which can capture the effect of inter die, intra gate variations, supply voltage(0.6-1.2 V) and temperature (0 - 100degC) on leakage. Results show that neural network based stack models can predict the PDF of leakage current across supply voltage and temperature accurately with the average error in mean being less than 2% and that in standard deviation being less than 5% across a range of voltage, temperature.

9 citations

##### Cited by

More filters

••

14 Oct 2017TL;DR: DRISA, a DRAM-based Reconfigurable In-Situ Accelerator architecture, is proposed to provide both powerful computing capability and large memory capacity/bandwidth to address the memory wall problem in traditional von Neumann architecture.

Abstract: Data movement between the processing units and the memory in traditional von Neumann architecture is creating the “memory wall” problem. To bridge the gap, two approaches, the memory-rich processor (more on-chip memory) and the compute-capable memory (processing-in-memory) have been studied. However, the first one has strong computing capability but limited memory capacity/bandwidth, whereas the second one is the exact the opposite.To address the challenge, we propose DRISA, a DRAM-based Reconfigurable In-Situ Accelerator architecture, to provide both powerful computing capability and large memory capacity/bandwidth. DRISA is primarily composed of DRAM memory arrays, in which every memory bitline can perform bitwise Boolean logic operations (such as NOR). DRISA can be reconfigured to compute various functions with the combination of the functionally complete Boolean logic operations and the proposed hierarchical internal data movement designs. We further optimize DRISA to achieve high performance by simultaneously activating multiple rows and subarrays to provide massive parallelism, unblocking the internal data movement bottlenecks, and optimizing activation latency and energy. We explore four design options and present a comprehensive case study to demonstrate significant acceleration of convolutional neural networks. The experimental results show that DRISA can achieve 8.8× speedup and 1.2× better energy efficiency compared with ASICs, and 7.7× speedup and 15× better energy efficiency over GPUs with integer operations.CCS CONCEPTS• Hardware → Dynamic memory; • Computer systems organization → reconfigurable computing; Neural networks;

179 citations

••

02 Jun 2018TL;DR: This paper presents the first proposal to enable scientific computing on memristive crossbars, and three techniques are explored — reducing overheads by exploiting exponent range locality, early termination of fixed-point computation, and static operation scheduling — that together enable a fixed- Point Memristive accelerator to perform high-precision floating point without the exorbitant cost of naïve floating-point emulation on fixed-pointers.

Abstract: Linear algebra is ubiquitous across virtually every field of science and engineering, from climate modeling to macroeconomics. This ubiquity makes linear algebra a prime candidate for hardware acceleration, which can improve both the run time and the energy efficiency of a wide range of scientific applications. Recent work on memristive hardware accelerators shows significant potential to speed up matrix-vector multiplication (MVM), a critical linear algebra kernel at the heart of neural network inference tasks. Regrettably, the proposed hardware is constrained to a narrow range of workloads: although the eight- to 16-bit computations afforded by memristive MVM accelerators are acceptable for machine learning, they are insufficient for scientific computing where high-precision floating point is the norm. This paper presents the first proposal to enable scientific computing on memristive crossbars. Three techniques are explored---reducing overheads by exploiting exponent range locality, early termination of fixed-point computation, and static operation scheduling---that together enable a fixed-point memristive accelerator to perform high-precision floating point without the exorbitant cost of naive floating-point emulation on fixed-point hardware. A heterogeneous collection of crossbars with varying sizes is proposed to efficiently handle sparse matrices, and an algorithm for mapping the dense subblocks of a sparse matrix to an appropriate set of crossbars is investigated. The accelerator can be combined with existing GPU-based systems to handle datasets that cannot be efficiently handled by the memristive accelerator alone. The proposed optimizations permit the memristive MVM concept to be applied to a wide range of problem domains, respectively improving the execution time and energy dissipation of sparse linear solvers by 10.3x and 10.9x over a purely GPU-based system.

35 citations

••

TL;DR: In this paper, a multiple-time programmable embedded non-volatile memory element, called the "charge trap transistor" (CTT), was proposed for high-$k$ -metal-gate CMOS technologies.

Abstract: The availability of on-chip non-volatile memory for advanced high- $k$ -metal-gate CMOS technology nodes has been limited due to integration and scaling challenges as well as operational voltage incompatibilities, while its need continues to grow rapidly in modern high-performance systems. By exploiting intrinsic device self-heating enhanced charge trapping in as fabricated high- $k$ -metal-gate logic devices, we introduce a unique multiple-time programmable embedded non-volatile memory element, called the ‘charge trap transistor’ (CTT), for high- $k$ -metal-gate CMOS technologies. Functionality and feasibility of using CTT memory devices have been demonstrated on 22 nm planar and 14 nm FinFET technology platforms, including fully functional product prototype memory arrays. These transistor memory devices offer high density ( $\sim 0.144\mu\mathrm{m}^{2}$ /bit for 22 nm and $\sim 0.082\mu\mathrm{m}^{2}$ /bit for 14 nm technology), logic voltage compatible and low peak power operation (~4mW), and excellent retention for a fully integrated and scalable embedded non-volatile memory without added process complexity or masks.

24 citations

••

TL;DR: An analog neural network computing engine based on CMOS-compatible charge-trap transistor (CTT) is proposed and obtained a performance comparable to state-of-the-art fully connected neural networks using 8-bit fixed-point resolution.

Abstract: An analog neural network computing engine based on CMOS-compatible charge-trap transistor (CTT) is proposed in this paper. CTT devices are used as analog multipliers. Compared to digital multipliers, CTT-based analog multiplier shows significant area and power reduction. The proposed computing engine is composed of a scalable CTT multiplier array and energy efficient analog–digital interfaces. By implementing the sequential analog fabric, the engine’s mixed-signal interfaces are simplified and hardware overhead remains constant regardless of the size of the array. A proof-of-concept 784 by 784 CTT computing engine is implemented using TSMC 28-nm CMOS technology and occupies 0.68 mm2. The simulated performance achieves 76.8 TOPS (8-bit) with 500 MHz clock frequency and consumes 14.8 mW. As an example, we utilize this computing engine to address a classic pattern recognition problem—classifying handwritten digits on MNIST database and obtained a performance comparable to state-of-the-art fully connected neural networks using 8-bit fixed-point resolution.

18 citations

••

TL;DR: Results show that the cumulative distribution function of leakage current of ISCAS'85 circuits can be predicted accurately with the error in mean and standard deviation, compared to Monte Carlo-based simulations, being less than 1% and 2% respectively across a range of voltage and temperature values.

Abstract: Artificial neural networks (ANNs) have shown great promise in modeling circuit parameters for computer aided design applications. Leakage currents, which depend on process parameters, supply voltage and temperature can be modeled accurately with ANNs. However, the complex nature of the ANN model, with the standard sigmoidal activation functions, does not allow analytical expressions for its mean and variance. We propose the use of a new activation function that allows us to derive an analytical expression for the mean and a semi-analytical expression for the variance of the ANN-based leakage model. To the best of our knowledge this is the first result in this direction. Our neural network model also includes the voltage and temperature as input parameters, thereby enabling voltage and temperature aware statistical leakage analysis (SLA). All existing SLA frameworks are closely tied to the exponential polynomial leakage model and hence fail to work with sophisticated ANN models. In this paper, we also set up an SLA framework that can efficiently work with these ANN models. Results show that the cumulative distribution function of leakage current of ISCAS'85 circuits can be predicted accurately with the error in mean and standard deviation, compared to Monte Carlo-based simulations, being less than 1% and 2% respectively across a range of voltage and temperature values.

18 citations