With India largely getting into the space of technology, information technology is one of its greatest assets. Not only this, the country also has long ago set its foot into the hardware world of computers.
Here are 10 top supercomputers that are developed in India, in alphabetical order.
1. Aaditya
Aaditya is an IBM iDataPlex Supercomputer, whose login and compute nodes are populated with two Intel Sandy Bridge 8-core processors. It has Intel Xeon Haswell E5-2670 2.6 GHz processors and total RAM storage of 15TB. There are 2384 compute nodes in the system. Red Hat Enterprise Linux is used as the operating system environment. Nodes are connected through infiniband interconnect technology.
Purpose: It is used by IITM, Pune for the purpose of research and development, providing weather conditions and simulating weather models for weather prediction.
Cost: ₹160 million
Speed: 719.2 Teraflops.
2. Anupam
Anupam is a series of supercomputers designed and developed by the Bhabha Atomic Research Centre (BARC) for their internal use. The latest in the series of ANUPAM systems is the ANUPAM-AGANYA supercomputer, inaugurated in the year 2016.
Purpose: The supercomputer series is used for molecular dynamical simulations, reactor physics, theoretical physics, computational chemistry, computational fluid dynamics, and finite element analysis.
Speed: 270 Teraflops
3. EKA
EKA is a supercomputer built by the Computational Research Laboratories (CRL) with hardware provided by Hewlett-Packard. This is developed by Tata Sons. It has a storage capacity of 40 TeraBytes and a memory of 28.7 TeraBytes. It was built within a short period of 6 weeks. At the time of its unveiling, it was the 4th fastest supercomputer in the world and the fastest in Asia. EKA follows a near-circular layout of the data centre unlike the traditional hot aisle and cold aisle rows. This near-circular layout enables the building of densely packed supercomputers, and this is the first time this architecture has been tried out on this scale. Eka has 15,000 processors (CPUs), needs 400 tonnes of cooling, and occupies 4,000 square feet of space. The project faced multiple challenges. For the first time in the world, fibre optic cables were used to connect the various CPUs.
Purpose: Research
Speed: 172.6 Teraflops
Cost: 118 crore
4. Mihir
Established at the National Center for Medium-Range Weather Forecast (NCMRWF), Noida, is a 2.8 PetaFlops unit at the NCMRWF. It consists of several computers that can deliver a peak power of 6.8 PetaFlops. Both Pratyush and Mihir were developed with an intention to move from the 165th position in the world and gain a position in the top 30s, in the space of supercomputers, of the “TOP500” list, released at the opening session of the 2013 International Supercomputing Conference in Leipzig, Germany.
Purpose: The purpose of Mihir, just like Pratyush, is to help in weather forecasting in the country. It has high-resolution models for the prediction on cyclones. Ocean state forecasts including marine water quality forecasts at very high resolution and natural calamities like Tsunami.
Maximum Speed: 6.8 PetaFlops
Cost: 450 crore
5. PARAM Ishan and PARAM Kanchenjunga
The latest supercomputers from the Param series, PARAM Ishan and the PARAM Kanchenjunga have been installed at IIT Guwahati, whereas PARAM Kanchenjunga is stationed at NIT Sikkim’s Superconducting Centre.
PARAM Kanchenjunga purpose: Computational chemistry, computational fluid dynamics, computational electromagnetics, civil engineering structures, nano-block self assemble, climate modelling and seismic data processing.
PARAM Kanchenjunga purpose: Engineering research conducted by the faculty and students at the institute as well as researchers across the state.
PARAM Ishan Speed: 250 TFLOPS
PARAM Kanchenjunga Speed: 15 TFLOPS
6. Param Shivay
In February this year, as a major step towards R&D excellence and innovation, PM Narendra Modi had inaugurated National Supercomputing Mission’s first indigenously built supercomputer called Param Shivay, at IIT BHU. It has the latest Intel based processor, high memory computer nodes and a peak computing power of 837 Teraflops. It is the most efficient supercomputer in the country. There are three supercomputing machines designed, manufactured and assembled at IIT Bhu, IISER Pune and IISER Kharagpur. This project is jointly implemented by the Department of Science and Technology (DST) and Ministry of Electronics and Information Technology (MeitY) and led by C-DAC and Indian Institute of Science (IISc), and supports the government’s vision of ‘Digital India’ and ‘Make in India’ initiatives.
Purpose: The purpose of this supercomputer is R&D. The National Supercomputing Mission was established to connect national academic and R&D institutions with a grid of over 70 high-performance computing facilities at an estimated cost of Rs 4,500 crore. Once operational, these high-performance computing facilities would improve weather services, disaster simulation and management, help faster processing of seismic data and help in computational biology.
Speed: 38.1 Teraflops
Cost: ₹ 32.5 crores
7. PARAM Yuva II
Made by the Centre for Development of Advanced Computing (CDAC), PARAM Yuva II is a supercomputer capable of delivering sustained performance of 360.8 TFLOPS on the community standard Linpack benchmark. It was inaugurated in the year 2013 and was made in a period of only three months at a cost of ₹160 million.
Purpose: The supercomputer is used for research in space, bioinformatics, weather forecasting, seismic data analysis, aeronautical engineering, scientific data processing and pharmaceutical development. Educational institutes can be linked to the computer through the national knowledge network.
Speed: 524 teraflops
Cost: ₹15 crores.
8. Pratyush
Pratyush is a supercomputer established at Indian Institute of Tropical Meteorology (IITM), Pune. Pratyush. The engineers of IITM, Pune worked under the leadership of Suryachandra A Rao and built Pratyush in 2018. The supercomputer, along with another one called Mihir, is currently the fastest supercomputer in the country and is the first multi-PetaFlops supercomputer ever built in India. The system is composed of 18 Compute cabinets and uses Cray’s Aries NOC with Dragonfly Interconnect network topology. It operates on Cray’s customised Linux OS, called Cray Linux Environment. The cluster supports architecture specific compilers from Cray as well as Intel and open-source GNU compilers. It is a 4.0 PetaFlops unit located at IITM, Pune. The unit, along with Mihir, gives a combined output of 6.8 PetaFlops.
Purpose: Pratyush is used in weather forecasting and climate monitoring. It helps in weather forecasting during monsoon, fishing, air quality, Tsunami, cyclones, earthquakes, lightning and other natural calamities such as floods and droughts. India is the fourth country in the world to have a High-Performance Computing facility dedicated for weather and climate research after Japan, the United States and the United Kingdom.
Maximum Speed: 6.8 PetaFlops
Cost: ₹438.9 Crore
9. Saga-220
Developed by ISRO, SAGA can perform at 450,000 gigaflops or 450 teraflops. It uses about 400 NVIDIA Tesla C2070 GPUs and 300 NVIDIA Tesla M2090 GPUs for acceleration. For processing, it uses 400 Intel QuadCore Xeon CPUs and 330 Intel HexCore Xeon CPUs. The storage capacity of this supercomputer is 120 TeraBytes. Over 6 months and with an investment of `14 crore it was designed and built by a team of 15 researchers at the Vikram Sarabhai Space Center, Thiruvananthapuram. It uses Mellanox’s InfiniBand network, NVIDIAs graphics processing units, and Intel’s central processing unit all supplied by Wipro. The supercomputer is loaded with open source software and ISROs own innovative complex aerospace application for the design and analysis of aerospace vehicle configurations.
Purpose: It is being used by space scientists at ISRO for solving complex aerospace problems and problems like fluid dynamic studies associated with building complex launch vehicles. It will help in reducing the design time as well as the computational cost to 1/15th of the equivalent CPU based system available in India.
Speed: 220 Teraflops
Cost: ₹14 crores
10. SahasraT (Cray XC40)
SahasraT supercomputer is located at Supercomputer Education and Research Centre (SERC), a facility at Indian Institute of Science (IISc). This Cray XC40 is a system that combines the capabilities of Intel’s latest Xeon Haswell processors for the CPU cluster and Nvidia’s K40 series of GPU cards and Intel Xeon-Phi Processor 7210 for the accelerator cluster connected using Crays’s own Aries high speed interconnect on a dragonfly topology with DDN’s high-performance storage units. It consists of Intel Haswell Xeon E5-2680v3 processors, NVIDIA K40 GPU accelerators and Intel Xeon Phi 5120D coprocessors and storage of 2.1 PetaBytes. There are around 1,500 processors and coprocessors and 44 GPUs to handle complex tasks in the system. SahasraT has been rated to 901.54 TFLOPS, which the highest rating amongst all supercomputers in India.
Purpose: Aerospace engineering, meteorology predictions and astrological simulations. It is also used for molecular and material research and mapping the entire climate condition of the particular region via simulation.
Speed: 1.46 petaflops
Cost: ₹82.70 crores
11. Vikram-100
Named after scientist Dr Vikram Sarabhai, Vikram-100 is a High-Performance Computing (HPC) Cluster.
It was inaugurated on 26 June 2015, by Prof. UR Rao at the Physical Research Laboratory (PRL). Vikram-100 has 97 compute nodes, each with two Intel Xeon E5-2670v3 12-core Intel Haswell CPUs at 2.30 GHz. It has a RAM size of 256 GB and local storage of 500 GB. 20 of these nodes also have two Nvidia Tesla K40 GPU cards each card capable of 1.66 Tflops. It has a storage capacity of 300 terabytes.
Purpose: Computing complex data in various areas like space and atmospheric sciences, geoscience, theoretical physics and solar physics. It can also help in numerical simulations.
Speed: 300 TeraBytes
Cost: ₹13 crore.
