January2025

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 The Thematic Unit of Excellence on Computational Materials Science (TUE-CMS), Solid State and Structural Chemistry Unit, Indian Institute of Science,Bangalore
System: The 140 node system consists 113 nodes with 125 GibiBytes RAM, 12 nodes with 503 GibiBytes RAM, 8 nodes with nVidia K40x GPU, 4 nodes with Intel 7120p Xeon Phi Coprocessor, 2 Master Nodes, 1 Login Node, 400 TB DDN Storage and Intel 12800 QDR Infiniband Switch.
Processor Socket: Intel Xeon E5-2630V4 @2.2 GHz dual twelve-core compute nodes) w/Infiniband
More Info
OEM:IBM, Bidder:HCL
3168 + 46080 + 492/2/
96.89 150.83
52 Rudra PoC,Pune
System: The Rudra PoC is a build with indigenously developed Rudra servers. Each densely designed server is 2U half-width form factor which has Intel Xeon Cascade Lake processors(dual socket, 6240R, 24C, 2.4GHz), two NVIDIA A100 (40GB), Mellanox HDR100 interconnect. This system is designed and implemented by the HPC Technologies Group, C-DAC under National Super Computing Mission (NSM). This PoC system is deployed at C-DAC, pune
Processor Socket: 2
More Info
OEM:C-DAC, Bidder:C-DAC
912//4
94.63 126.88
53 Inter-University Centre for Astronomy and Astrophysics (IUCAA),Pune
System: Pegasus supercomputer is of 159.72 TF and built with 60 nodes cluster using Intel Gold 6142 Processor,1 PF of PFS Storage with 15Gbps performance, EDR interconnect, IBM Spectrum cluster manager and PBS Pro Job scheduler. Compute hardware is on Lenovo make.
Processor Socket: is of 159.72 TF and built with 60 nodes cluster using Intel Gold 6142 Processor,1 PF of PFS Storage with 15Gbps performance, EDR interconnect, IBM Spectrum cluster manager and PBS Pro Job scheduler. Compute hardware is on Lenovo make.
More Info
OEM:Lenovo, Bidder:Locuz
1920//
93.9 156
54 S.N.Bose National Institute of Basic sciences,Kolkata
System:
Processor Socket: each with 2 X Intel SKL 6148 Processor, 12 X 16 GB 2666MHz Memory with 2 dedicated Boot NodesCray Aries Interconnect,Cray XC-50 Air Cooled System with 48 x Dual Processor Compute Nodes, with Skylake Gold 6148 2.4Ghz CPUs 192GB Per Node Memory in 44 Nodes and 768GB per Node Memory in 4 nodes. Plus Additional 8 Service nodes. Interconnect is Cray Proprietary ARIES Interconnect.
More Info
OEM:HPC, Bidder:Cray
1920//
91.96 147.45
55 PARAM Rudra C-DAC Delhi,Delhi
System: The PARAM Rudra is built with indigenously developed Rudra servers by C-DAC. It is a heterogeneous and hybrid configuration of Intel Xeon Cascade Lake 2nd Gen processors, NVIDIA A100 80GB, Mellanox HDR interconnect with C-DAC HPC Software stack having peak capacity of 200 TF. This system is designed and implemented by the HPC Tech Group, C-DAC at C-DAC Delhi.
Processor Socket: 2
More Info
OEM:C-DAC, Bidder:NA
1824/2/38
89.13 140.06
56 Tata Consultancy Services,Bangalore
System: 176 Nodes, 2816 Cores, 4.2TB Total nodes Memory, Gluster Parallel file system, with 88TB Usable storage Space.
Processor Socket: Intel Xeon E5-2680v3 @2.5 GHz dual twelve-core compute nodes) w/Infiniband
More Info
OEM:IBM, Bidder:Lenovo
2688/2/112
86.09 107.52
57 Inter-University Centre for Astronomy and Astrophysics (IUCAA),Pune
System: Two numbers of Apollo k6000 Chassis, each having 24 * ProLiant XL230k Gen10 compute node with dual CPUs (Intel(R) Xeon(R) Gold 6142 CPU @ 2.60GHz ) running Scientific Linux Operating System Version 7.5, and connected by 10 G ETH.
Processor Socket: Intel(R) Xeon(R) Gold 6142 CPU @ 2.60GHz ) running Scientific Linux Operating System Version 7.5, and connected by 10 G ETH.
More Info
OEM:HPE, Bidder:Concept
1536//
80.06 127.97
58 Jawaharlal Nehru Centre for Advanced Scientific Research,Bangalore
System: The HPC-cluster at JNCASR contains 2 master nodes ,1 login node , 108 compute nodes,1 spare node , 1 iml server, 3 GPU nodes, 2 lustre MDS nodes, 2 lustre OSS nodes with 200TB of usable PFS storage configured using NetApp E5500 - hardware RAID storage array with dual controllers each with 12GB Cache and dual port IB interfaces and 168 x 2TB 7.2K RPM NL-SAS disk drives.
Processor Socket: Intel Xeon E5-2680v3 @ 2.5 GHz dual twelve-core processor nodes) w/Infiniband
More Info
OEM:Lenovo/IBM, Bidder:Wipro
2592/2/
77.13 103.68
59 Indian Institute of Technology, Madras,Chennai
System: The Virgo cluster is a IBM System x iDataPlex cluster with 292 dx360 M4 Server Intel Xeon Processor E5-2670 based nodes as Compute Nodes, 4672 cores, with SLES 11 SP2 OS, 19.13 Terabytes of total memory, w/Infiniband FDR, fully non-blocking FAT tree. 150 TB Usable Capacity Storage based on IBM General Parallel File System over 2 x IBM DS3534 Storage System. The cluster also has 2 master nodes and 4 storage nodes based on Intel Xeon Processor E5-2670 processors with 65 GB memory per node. The storage nodes are based on IBM DS3524 Dual Controller Storage Subsystem.
Processor Socket: Intel(R) Xeon(R) CPU E5-2670 @ 2.6 GHz dual octo-core processor nodes w/Infiniband
More Info
OEM:IBM, Bidder:SBA
4672/2/
75.5 97.2
60 IIT,Jodhpur
System: Lenovo ThinkSystem 20x SR645, 2 x AMD EPYC 7742 2.25GHz, 64C, 16 x 16GB DDR4 3200MHz, 2 x 480GB SSD, 100G Mellanox IB Adapter, Redundant Power supply, 400TB IBM Spectrum Scale PFS
Processor Socket: 2
More Info
OEM:Lenovo, Bidder:Access Data Systems
2560//20
71.09 92.16