Historically, India has struggled to establish an advantage in supercomputing, but its government is looking to change that with the announcement of a slew of new supercomputers and homegrown technologies, including its server platform” Rudra”. India’s Ministry of Science and Technology announced this week that it plans to deploy nine new supercomputers this year as a continuation of India’s National Supercomputing Mission, which since 2015 has been seeking to install a large number of supercomputers in the country with a steadily increasing percentage of indigenous components.
The nine new supercomputers, the ministry said, will be commissioned and installed in the coming year at institutes such as IIT (Indian Institute of Technology) Bombay, IIT Madras, IIT Patna, IIT Delhi, Inter-University Accelerator Center (IUAC), Center for Development of Advanced Computing (C-DAC) Pune, SN Bose National Center for Basic Sciences (SNBNCBS), National Center for Radio Astrophysics (NCRA ) and the National Informatics Center (NIC).
The news came below the fold in an announcement that primarily sought to highlight the National Supercomputing Mission’s current wins: namely, 10 institutions that have already installed supercomputing infrastructure under the plan. These institutions include Indian Institute of Science (IISc), IISER (Indian Institute of Science Education and Research) Pune, Jawaharlal Nehru Center for Advanced Scientific Research (JNCASR), National Agri-Food Biotechnology Institute (NABI) and various C-DACs and IITs. The ministry added that “final stage installation work is underway at five other facilities.”
Altogether it appears: 15 systems with nine more apparently on the way. The ministry says this will allow hundreds of institutions and thousands of researchers to access HPC facilities.
Yet this remains a bit short of the original goals of the National Supercomputing Mission, which was funded to the tune of $600 million and, when launched in 2015, aimed to deploy 73 supercomputers in India by 2022, or about three times the projected new total in the coming year.
Currently, India has three public supercomputers powerful enough to rank in the Top500: the PARAM Siddhi-AI, built by Atos, installed in 2018 at C-DAC Pune (4.6 petaflops Linpack, #102); and the HPE-built Pratyush and Mihir systems, both installed in 2018 at the Indian Institute of Tropical Meteorology (#121 at 3.8 petaflops Linpack and #228 at 2.6 petaflops Linpack, respectively).
The ministry also pointed out that the design and development of indigenous server nodes, interconnect switches, storage technologies and software stacks for HPC continues with “85% indigenous manufacturing”, suggesting that these technologies are intended for use in future supercomputers. As examples, they highlighted India’s “first native server platform” (called Rudra) and a “next-generation native HPC interconnect” (called Trinetra). Rudra, detailed in December, is a dual-socket Intel server that supports GPU acceleration (see graphic above).
India’s new emphasis on its national supercomputing mission and local HPC design and manufacturing is another signal that, as Eric Eppe of Atos put it, “the pandemic [has] reminded everyone of the importance of being sovereign. Initiatives such as the European EuroHPC, the European Processor Initiative and the European Chips Act have increasingly sought to internalize elements of the supercomputing supply and operations chains that have historically been outsourced to world leaders in HPC such as the United States and China.
Comments are closed.