PDPravin A. Dwaramwar

Pravin A. Dwaramwar

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

Pravin A. Dwaramwar has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Exploring the Cultural Impacts of Artificial Intelligence on Sustainable Entrepreneurship Development

    by Karthikeyan Murugan, Archit Dev, Krishna Kumari, Pravin A. Dwaramwar, R. Somasundaram, S. Boopathi

    The chapter discusses the connection between artificial intelligence (AI) and sustainable entrepreneurship, emphasizing the need for understanding its sociological and cultural implications. It emphasizes the significance of cultural sensitivity in AI development and adoption, highlighting how it can help organizations mitigate biases and foster inclusive AI ecosystems. While AI technologies can improve efficiency, innovation, and scalability, they also raise privacy, autonomy, and job displacement concerns, necessitating a balance between technological advancements and ethical considerations. It calls for interdisciplinary collaboration between sociologists, HR professionals, entrepreneurs, and technologists to develop AI-driven sustainable entrepreneurship initiatives. By incorporating sociocultural insights, organizations can harness AI's transformative potential for sustainable entrepreneurship.


  • AI-Driven Energy Optimization in High-Performance Computing

    by S. Subashree, T. Akila, Pravin A Dwaramwar, Saurabh Chandra, Ketki P. Kshirsagar

    The rapid digital transformation in this era has led to the rise of high-performance computing (HPC) systems as a hub for addressing computationally intensive requirements, but their energy consumption is also a growing concern due to sustainability efforts. This chapter surveys the potential transformative impact of Artificial Intelligence (AI) in optimizing energy management in the context of HPC systems. With machine learning algorithms and predictive analytics, an AI system can dynamically adapt to power use levels, optimize workload distributions, and reduce energy consumption without compromising computational performance. The techniques discussed in this chapter include AI-driven techniques founded on computational methods such as reinforcement learning and neural network models that change HPC workloads appropriately to support real-time decision-making and energy-efficient operations. This chapter presents successful AI-based system implementations while highlighting high performance and energy savings.


  • Architectures of High-Performance Computing Systems for Machine Learning Workloads

    by M. Raju, Kimsy Gulhane, Banu A. Gulshan, Saurabh Chandra, Pravin A. Dwaramwar, Sampath Boopathi

    This chapter explores the architectures of high-performance computing (HPC) systems designed for machine learning (ML) workloads with special attention paid to advanced hardware and software optimizations intended to accelerate computational efficiency. It discusses the state-of-the-art use of GPUs, TPUs, and FPGAs in parallelizing operations, optimizing deep learning models during training and inference periods. The final section covers distributed computing frameworks, such as Apache Spark and Hadoop, that also provide support for big data processing in clusters. It also examines the challenges arising from the handling of data, distribution of resources, and scalability aspects as well as emerging solutions, such as memory-centric architectures and quantum computing, respectively. This chapter provides a comprehensive understanding of the evolution of HPC systems to meet the requirements of modern ML workloads, faster model development, reduced energy consumption, and better scalability, along with some of the trends and future research directions in HPC for ML.