Sreenivas Subramoney

Sreenivas Subramoney

Auto-generated profile, based on publicly available information

Sreenivas Subramoney

Intel, Bengaluru, India

Research Focus & Metrics

Academic Career Length

27Years (2000-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2019 in APL Materials
Verified
Journal Article
Open access
PDF file
DKDevesh Khilwani
Devesh Khilwani
VMVineet Moghe
Vineet Moghe
SLSandip Lashkare
Sandip Lashkare
VSVivek Saraswat
Vivek Saraswat
PKPankaj Kumbhare
Pankaj Kumbhare
UGUdayan Ganguly
Udayan Ganguly
The neural network enables efficient solutions for Nondeterministic Polynomial-time (NP) hard problems, which are challenging for conventional von Neumann computing. The hardware implementation, i.e., neuromorphic computing, aspires to enhance this efficiency by custom hardware. Particularly, NP hard graphical constraint optimization problems are solved by a network of stochastic binary neurons to form a Boltzmann Machine (BM). The implementation of stochastic neurons in hardware is a major challenge. In this work, we demonstrate that the high to low resistance switching (set) process of a PrxCa1−xMnO3 (PCMO) based RRAM (Resistive Random Access Memory) is probabilistic. Additionally, the voltage-dependent probability distribution approximates a sigmoid function with 1.35%–3.5% error. Such a sigmoid function is required for a BM. Thus, the Analog Approximate Sigmoid (AAS) stochastic neuron is proposed to solve the maximum cut—an NP hard problem. It is compared with Digital Precision-controlled Sigmoid (DPS) implementation using (a) pure CMOS design and (b) hybrid (RRAM integrated with CMOS). The AAS design solves the problem with 98% accuracy, which is comparable with the DPS design but with 10× area and 4× energy advantage. Thus, ASIC neuro-processors based on novel analog neuromorphic devices based BM are promising for efficiently solving large scale NP hard optimization problems.

15 February 2024 in ACM Transactions on Architecture and Code Optimization
Verified
Journal Article
Open access
CFCan Firtina
Can Firtina
KPKamlesh Pillai
Kamlesh Pillai
GKgurpreet singh kalsi
gurpreet singh kalsi
BSBharathwaj Suresh
Bharathwaj Suresh
DCDamla Senol Cali
Damla Senol Cali
OMOnur Mutlu
Onur Mutlu
Profile hidden Markov models (pHMMs) are widely employed in various bioinformatics applications to identify similarities between biological sequences, such as DNA or protein sequences. In pHMMs, sequences are represented as graph structures, where states and edges capture modifications (i.e., insertions, deletions, and substitutions) by assigning probabilities to them. These probabilities are subsequently used to compute the similarity score between a sequence and a pHMM graph. The Baum-Welch algorithm, a prevalent and highly accurate method, utilizes these probabilities to optimize and compute similarity scores. Accurate computation of these probabilities is essential for the correct identification of sequence similarities. However, the Baum-Welch algorithm is computationally intensive, and existing solutions offer either software-only or hardware-only approaches with fixed pHMM designs. When we analyze state-of-the-art works, we identify an urgent need for a flexible, high-performance, and energy-efficient hardware-software co-design to address the major inefficiencies in the Baum-Welch algorithm for pHMMs.We introduceApHMM, thefirstflexible acceleration framework designed to significantly reduce both computational and energy overheads associated with the Baum-Welch algorithm for pHMMs. ApHMM employs hardware-software co-design to tackle the major inefficiencies in the Baum-Welch algorithm by (1) designing flexible hardware to accommodate various pHMM designs, (2) exploiting predictable data dependency patterns through on-chip memory with memoization techniques, (3) rapidly filtering out unnecessary computations using a hardware-based filter, and (4) minimizing redundant computations.ApHMM achieves substantial speedups of 15.55×–260.03×, 1.83×–5.34×, and 27.97× when compared to CPU, GPU, and FPGA implementations of the Baum-Welch algorithm, respectively. ApHMM outperforms state-of-the-art CPU implementations in three key bioinformatics applications: (1) error correction, (2) protein family search, and (3) multiple sequence alignment, by 1.29×–59.94×, 1.03×–1.75×, and 1.03×–1.95×, respectively, while improving their energy efficiency by 64.24×–115.46×, 1.75×, and 1.96×.

13 June 2025
Verified
Conference Paper
Open access
SSSathvik Swaminathan
Sathvik Swaminathan
SKSandeep Kumar
Sandeep Kumar
APAravinda Prasad
Aravinda Prasad
SSSreenivas Subramoney
Sreenivas Subramoney

17 December 2022
Verified
Book Chapter
Closed access
PDF file
ALAshwin Lele
Ashwin Lele
SJSrivatsava Jandhyala
Srivatsava Jandhyala
SGSaurabh Gangurde
Saurabh Gangurde
VSVirendra Singh
Virendra Singh
SSSreenivas Subramoney
Sreenivas Subramoney
UGUdayan Ganguly
Udayan Ganguly

26 April 2026
Verified
Conference Paper
Open access
SKSandeep Kumar
Sandeep Kumar
APAravinda Prasad
Aravinda Prasad
SSSreenivas Subramoney
Sreenivas Subramoney
Tiered memory systems are the norm to effectively tackle the increasing memory total cost of ownership (TCO) in modern data centers. We propose TierScape to tame memory TCO through the novel creation and judicious management of multiple compressed memory tiers along with multiple byte-addressable tiers. As opposed to conventional tiering solutions that employ a single compressed memory tier, we harness multiple compressed tiers implemented through a combination of different compression algorithms, memory allocators for compressed objects, and backing media to store compressed objects. These compressed tiers represent distinct points in the access latency, data compressibility, and unit memory usage cost spectrum, allowing rich and flexible trade-offs between memory TCO savings and application performance impact. A key advantage with TierScape is that it enables aggressive memory TCO saving opportunities by placing warm data in low latency compressed tiers with a reasonable performance impact while simultaneously placing cold data in the best memory TCO saving tiers. TierScape presents a comprehensive, rigorous, and tunable analytical cost model for performance and TCO trade-off based on continuous monitoring of the application's data access profile. Guided by this model, TierScape takes informed actions to dynamically manage the placement and migration of application data across multiple compressed and byte-addressable tiers. We believe TierScape represents an important server system configuration and optimization capability to achieve the best SLA-aware performance per dollar for applications hosted in production data center environments. For a set of real-world benchmarks, TierScape reduces memory TCO by 15.1–23.6% percentage points while maintaining performance parity or improves performance by 2.61–10.0% percentage points while maintaining memory TCO parity compared to state-of-the-art tiering solutions.

4 December 2021 in Mathematics
Verified
Journal Article
Open access
PDF file
BSBharathwaj Suresh
Bharathwaj Suresh
KPKamlesh Pillai
Kamlesh Pillai
GKgurpreet singh kalsi
gurpreet singh kalsi
AAAvishaii Abuhatzera
Avishaii Abuhatzera
SSSreenivas Subramoney
Sreenivas Subramoney
Deep Neural Networks (DNNs) have set state-of-the-art performance numbers in diverse fields of electronics (computer vision, voice recognition), biology, bioinformatics, etc. However, the process of learning (training) from the data and application of the learnt information (inference) process requires huge computational resources. Approximate computing is a common method to reduce computation cost, but it introduces loss in task accuracy, which limits their application. Using an inherent property of Rectified Linear Unit (ReLU), a popular activation function, we propose a mathematical model to perform MAC operation using reduced precision for predicting negative values early. We also propose a method to perform hierarchical computation to achieve the same results as IEEE754 full precision compute. Applying this method on ResNet50 and VGG16 shows that up to 80% of ReLU zeros (which is 50% of all ReLU outputs) can be predicted and detected early by using just 3 out of 23 mantissa bits. This method is equally applicable to other floating-point representations.