Prime Minister Narendra Modi on September 26 inaugurated three PARAM Rudra supercomputers worth Rs 130 crore, including this one at the S. N. Bose National Centre for Basic Sciences (SNBNCBS). (Photo - Twitter/@snbkol)Prime Minister Narendra Modi on September 26 inaugurated three PARAM Rudra supercomputers worth Rs 130 crore.
Describing the “momentous” occasion, the Prime Minister said that the supercomputers will help provide “state-of-the-art advanced facilities to the scientific community” in India, and that it will help advance research in fields ranging from “physics and cosmology to earth sciences”.
Developed indigenously under the National Supercomputing Mission (NSM), the three new supercomputers are located in Pune, Delhi, and Kolkata, and make up a High-Performance Computing (HPC) system tailored for weather and climate research.
What are supercomputers? And why is India trying to indigenously develop these expensive machines?
A supercomputer is a large computing system specifically designed to solve complex, scientific and industrial challenges, which tend to be time-consuming and computation-intensive.
They are used in quantum mechanics, weapons research, weather forecasting and climate research, oil and gas exploration, molecular dynamics and physical simulations, data analytics and big data — all of which require a high computing capacity which are unavailable with regular systems.
Supercomputers are expensive systems that take up a large room worth of space in the form of multiple rows with racks holding computer nodes with many cores (see adjoining image). A High Performance Computing (HPC) system is made up of several such supercomputers put together.
A general view of IBM Blue Gene P supercomputer, one of the largest examples of parallel computing as of year 2023.
Launched in 2015, the NSM aimed to build a grid of 70 powerful supercomputing systems for use in research and development centres, and higher education institutions. These supercomputers were later networked on the National Supercomputing Grid over the National Knowledge Network (NKN).
The seven-year mission, with an economic outlay of Rs 4,500 crores, is jointly spearheaded by the Department of Science and Technology (DST) and the Ministry of Electronics and Information Technology (MeitY). The Centre for Development of Advanced Computing (CDAC) and Indian Institute of Science (IISc), Bengaluru act as the nodal implementing agencies. Two phases of the NSM are complete, and the third phase is now underway.
Some of the broad applications of the NSM include climate modelling, weather prediction, aerospace engineering, computational biology, molecular dynamics, atomic energy simulations, national security and defence applications, seismic analysis, disaster simulations and management, computational chemistry, computational material science and nanomaterials, astrophysics, large complex systems simulations, cyber-physical systems, big data analytics, finance, and information repositories.
India bore the brunt of technology-denial by Western nations on two significant occasions. The first delayed India’s forays into space in the 1970s, and the second stopped it from building supercomputers in the 1990s. However, this further motivated India to develop pocket-friendly, indigenous technologies in the decades that followed.
The NSM is born out of this ambition to develop indigenous technologies, and gave a much needed boost to India’s computational capacity — something that PM Modi likened to “a tool to be used as soft power” last week.
Since the launch of the programme, more than 20 supercomputing systems have been deployed nationwide (See Table). All these broadly support research in bioinformatics, engineering, disaster simulation and management, material modelling, quantum chemistry, weather, ocean and climate, astronomy, material science, energy, medical research and for the smooth operations of the Micro, Small and Medium Enterprises sector.
| Supercomputer | Institute | Capacity and memory |
| PARAM Shivay | IIT BHU | 837 teraFLOPS and a total memory of 54.5 TB |
| PARAM Shakti | IIT Kharagpur | 1.66 petaFLOPS and a total memory of 103.125 TB |
| PARAM Brahma | Indian Institute of Science Education and Research, Pune | 1.75 petaFLOPS and a total memory of 56.8 TB |
| PARAM Yukti | Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore | 1.8 petaFLOPS and a total memory of 52.416 TB |
| PARAM Sanganak | IIT Kanpur | 1.67 petaFLOPS and a total memory of 104.832 TB |
| PARAM Pravega | Indian Institute of Science, Bangalore | 3.3 petaFLOPS and a total memory of 245.945 TB |
| PARAM Seva | IIT Hyderabad | 838 teraFLOPS and a total memory of 52.416 TB |
| PARAM Smriti | National Agri-Food Biotechnology Institute, Mohali | 838 teraFLOPS |
| PARAM Utkarsh | CDAC, Bangalore | 838 teraFLOPS and a total memory of 52.416 TB |
| PARAM Ganga | IIT Roorkee | 1.66 petaFLOPS and a total memory of 104.832 TB |
| PARAM Ananta | IIT Gandhinagar | 838 teraFLOPS and a total memory of 52.416 TB |
| PARAM Porul | NIT, Trichy | 838 teraFLOPS |
| PARAM Himalaya | IIT Mandi | 838 teraFLOPS and a total memory of 52.416 TB |
| PARAM Kamrupa | IIT Guwahati | 838 teraFLOPS and a total memory of 52.416 TB |
| PARAM Siddhi | AI CDAC, Pune | 5.2 petaFLOPS and 210 petaFLOPS (AI) |
| PARAM Rudra | Giant Metrewave Radio Telescope, Pune | 1 petaFLOPS |
| Inter-University Accelerator Centre, Delhi | 838 teraFLOPS | |
| SN Bose National Centre for Basic Sciences, Kolkata | 838 teraFLOPS |
Here are some notable facts about the NSM.
* During 2019-2023, a total capacity of 24.83 petaFLOPS HPC machines were commissioned.
* 1.75 lakh people were trained in HPCs, and 5,930 experts from over 100 institutes used the newly-developed facilities. Over 73.25 lakh high performance computational queries were executed.
* During this period, seven systems offering computing power of above one petaFLOPS, eight systems offering computing capacities ranging between 500 teraFLOPS-1 petaFLOPS and 13 systems with capacities ranging between 50 teraFLOPS-500 teraFLOPS were installed.
* R&D systems at various centres of CDAC: The SANGAM Testbed, PARAM Shrestha, PARAM Embryo, PARAM Neel, PARAM Spoorthi, PARAM Sampooran, Bioinformatics facility and system software lab are presently operational.
* Aimed at training personnel in high performance computational skills, the mission established dedicated learning centres functioning at IIT Kharagpur, Palakkad, Chennai, Goa and CDAC, Pune, which have been operating PARAM Vidya (1 to 5 versions).


