Top Supercomputers- India

January2023

Following is the ranking of the systems in terms of Rmax (Linpack Benchmark performance).

C - CPUs, G-GPUs, ICO - Intel Co-processors. Cores (4th column) that are not qualified with C/G/ICO refer to CPUs.
OEM - Original Equipment Manufacturer, SI - System Integrator.

Rank	Site	System	Core/Processor/Socket/Nodes	Rmax (TFlops)	Rpeak (TFlops)
51	Aeronautical Development Agency, DRDO,Bangalore	Fujitsu Cluster PRIMERGY CX250 S2 cluster based on Dual Intel Xeon E5-2697 V2 twelve core processor at 2.7 GHz w/Infiniband FDR, fully non-blocking FAT tree. Total memory of 12.28 TB, and total storage 150 TB based on IBRIX-Parallel File System. Connected by Mellanox Infiniband system.Intel Xeon E5-2697v2 @ 2.7 GHz dual twelve-core compute nodes) w/ Mellanox FDR Infiniband More Info OEM:Fujitsu, Bidder:TATA Elxi	3072/2/	60.55	66.35
52	S.N. Bose National Centre for Basic Sciences,Kolkata	SNB Cray XE6 Cluster based on AMD Opteron 6300 series (Abu Dhabi) processors, 244 nodes, 7808 cores, with total memory of 8.39 TB, CRAY LINUX ENVIRONMENT ( CLE) Compute Node Linux operating system, 250 TB total usable storage with Lustre file system, nodes connected by Cray Gemini interconnect.AMD Opteron 6300 Series (ABU DHABI) @2.4 GHz dual sixteen-core processor nodes w/Crays Gemini primary interconnect More Info OEM:Cray, Bidder:Cray	7808/2/	60	74.95
53	Physical Research Laboratory,Ahmedabad	IBM/Lenovo NeXtScale System- M5 based Servers, Xeon E5-2670v3 12C 2.3 GHz, Nvidia K40 GPU, Infiniband FDR Interconnect with 300TB GPFS parallel file system. The system having 2328 CPU Cores/194 CPU Socket/77 Pure CPU Compute Node and 11520 GPU Cuda Cores/40 GPU/20 Hybrid(CPU+GPU)Compute Nodes.Intel Xeon Processor E5-2670 v3 @ 2.355 GHz dual twelve-core processor CPU nodes. Twenty nodes also contain dual 2880-core NVIDIA K40 GPUs) w/ 4X FDR Infiniband primary interconnect More Info OEM:Lenovo, Bidder:TCS	2328 + 115200/2/	55.6	68
54	Indian Institute of Tropical Meteorology,Pune	Prithvi HPC is built on IBM P6 575 nodes totaling 117 numbers including 4 GPFS nodes and 3 Login nodes. Each node is populated with 32 cores of IBM P 6 CPU running at 4.7 G Hz. The total number of cores works to 3744 cores. The Prithvi HPC has Interconnectivity using Infiniband Switches and Ethernet switches for Management purposes. 15.34 TB total memory. Total of 2800 TB of IBM DS Storage including Online, Nearline and Archival Storage. The cluster also has GPFS and other Managment SW.IBM P6 4.7 GHz sixteen dual-core processor 575 nodes w/Infiniband 4X More Info OEM:IBM, Bidder:HCL	3744/2/	45.84	66.18
55	IGIB Delhi,Delhi	The system consists of Cascade lake 6230 , 24 nodes each with 2 sockets (2 x20 core) interconnect with IB Switch HDR 100.2 More Info OEM:Lenovo, Bidder:Lenovo	960/2/24	44.81	64.5
56	TATA INSTITUTE OF FUNDAMENTAL RESEARCH,Hyderabad	Intel Xeon Phi cluster, 68 CPU nodes, 1360 CPU cores, 4 GPU nodes of a total of 16 NVIDIA K40 GPUs, Cent OS Linux 7.2, 4.45 TB total memory, with Infiniband interconnect.Intel Xeon E5-2630V4 @2.2 GHz dual ten-core compute nodes) w/Infiniband More Info OEM:Supermicro, Bidder:Netweb	1360 + 46080/2/	43.59	70.85
57	Variable Energy Cyclotron Centre (VECC),Kolkata	Dell PowerEdge FX2 Enclosure 2U Chassis. --- 12Nos Server Nodes - Dell™ PowerEdge™ PowerEdge FC630 Server Nodes --- 48Nos 02Nos x Intel Xeon E5-2680 v4/14C/ 2.4GHz,35M Cache,9.60GT/s QPI,Turbo,HT,14C/28T (120W) 800GB Solid State Drive 128GB RAM Total CPU : 96NosPowerEdge FC630 Server Nodes --- 48Nos 02Nos x Intel Xeon E5-2680 v4 2.4GHz,35M Cache,9.60GT/s QPI,Turbo,HT,14C/28T (120W) , 128GB More Info OEM:Dell, Bidder:Micropoint	1344//	43.1	51.9
58	Research Centre Imarat,Hyderabad	RCI-PGAD-Advanced Numerical Simulation for enabling Research ( ANSER )Intel Xeon Processor E5-2670 2.6GHz More Info OEM:HP, Bidder:HP	2432/2/	41.95	50.58
59	Institute of Physics(IOP),Bhubaneswar	Fujitsu PY CX400 M1 (20 Nodes with dual E5-2680 V3 @ 2.5Ghz processor, 128 GB RAM, 2 Nos. Intel Co-Processor 7120P Total XEON PHI : 40Nos , Fujitsu E5-2680 V3 @ 2.5Ghz processor, 128 GB RAM, 2 Nos. Nvidia Tesla K8020 Nodes with dual E5-2680 V3 @ 2.5Ghz processor, 128 GB RAM, 2 Nos. Intel Co-Processor 7120P,60 CPU Nodes, 40 GPU-K80 More Info OEM:Fujitsu, Bidder:Micropoint	//	41.79	48
60	High Energy Materials Research Laboratory (HEMRL),Pune	Fujitsu Server PRIMERGY RX2530 M2 One Master node with 54 Compute Nodes/12core per Sockets with Dual E5-2680 V4@ 2.10Ghz processor 128GB RAM connected with Intel OmniPath 100Gbpsis 53.76 TF cluster, built with 20 CPU nodes and 4 GPU nodes, Using Xeon-G 6230 Processor.PFS storage with 220 TB Lustre based. Primary interconnect is Infiniband. More Info OEM:HPE, Bidder:Concept	800//	39.5	53.76

,