CCF - HPC | MANIT Bhopal

Coming Soon

CCF: High Performance Cluster

The High Performance Computing (HPC) Facility at MANIT Bhopal is a robust, multi-purpose Linux environment engineered to tackle complex computational challenges across the academic community. Powered by high-speed Xeon processors and seamless gigabit internode connectivity, the cluster provides the raw power necessary for data-intensive research and advanced teaching through two specialized environments: CPU compute nodes optimized for high-throughput parallel processing and high-end NVIDIA GPU nodes designed for intensive AI/ML and parallel workloads. By offering these high-performance resources to all members of the institute, the facility serves as a vital hub for accelerating innovation, discovery, and large-scale scientific modeling.

High Performance Cluster is equipped with the following components:

Racks:

HPC is housed with 2x dedicated air-cooled smart racks containing of 8 compute nodes, 2 GPU nodes, 1 Nvidia DGX A100 Box, 2 Master/Login Nodes, 4 I/O Nodes and 1 Management Node. Cluster also features a Dell EMC Storage.

Compute Nodes:

Each of the 8 compute nodes is a Dell PowerEdge R640 Server equipped with 2 x Intel Xeon® Gold 6248 20-core CPUs i.e. 40 Cores running at 2.50 GHz and 192 GB of memory.

GPU Nodes:

Each of the 2 GPU Nodes is a Dell PowerEdge R740 Server equipped with 2 x Intel Xeon® Gold 6248 20-core CPUs i.e. 40 Cores running at 2.50 GHz, 192 GB of memory, and a Nvidia Quadro RTX 6000 with 22 GB of GPU memory.

DGX A100:

The DGX node is a universal system for all AI workloads, from analytics to training to inference, built by Nvidia technologies. The system has dual AMD ROME 7742 128-Cores total at 2.25 GHz (base) and 3.4 GHz (max boost) of processing power along with 1 TB of system memory.

Master/Login Nodes:

Each of the 2 Master/Login nodes is a Dell PowerEdge R740 Server equipped with 2 x Intel Xeon® Gold 6248 20-core CPUs i.e. 40 Cores running at 2.50 GHz and 192 GB of memory.

I/O Nodes:

Each of the 4 I/O nodes is a Dell PowerEdge R740 Server equipped with 2 x Intel Xeon® Gold 6248 20-core CPUs i.e. 40 Cores running at 2.50 GHz and 192 GB of memory.

Management Node:

The Management node is a Dell PowerEdge R640 Server equipped with 2 x Intel Xeon® Silver 4214 12-core CPUs i.e. 24 Cores running at 2.20 GHz and 192 GB of memory.

Storage:

The cluster is equipped with Dell EMC Storage using the Lustre parallel file system, offering a high performance and a usable storage capacity of 122 TB approx.

Interconnects:

HPC is interconnected with a 100 Gbps network through 2 x Mellanox SB7800 36 Port Infiniband switches, and a 1 Gbps network via Dell S3048 1 x 48-port Gigabit Ethernet and 4 x 4-port Ten-Gigabit switch.