Parallel computing is a form of computation in which many calculations are carried out simultaneously, operating on the principle that large problems can often be divided into smaller ones, which are then solved concurrently ("in parallel").
In Computer Center, IITK There are few large scale, high performance, parallel processing compute servers available here, which are designed for speed, scalability, flexibility, manageability and availability.
HPC2013 Cluster |
|
| Host Name | hpc2013.hpc.iitk.ac.in |
| Configuration | 901 Nodes Cluster with Intel Xeon E5-2670v2 10C 2.5GHz; 17980/20 per node (vector) Cores with Rmax of 344 TFlops and Rpeak of 359 Tflops, Infiniband - FDR (56Gb/S) {Mellanox} 36P (edge Switch) 2 * 648 P (Single Chasis Director Switch) with 513 T (DDN) Storage. |
| Operating System | RedHat Enterprise Linux (RHEL) 6.5 |
| Specific Usage | High Performance Computing with Several Softwares |
HPC2010 Extension (Sandy Bridge) Cluster |
|
| Host Name | hpc2010.hpc.iitk.ac.in |
| Configuration | 368+96 Nodes Cluster with Intel Xeon 2.93Ghz 2.6Ghz; 2944/8 (non-vector) + 1536/16 (vector) Cores with Rmax of 29 TTflops ~= 25 Tflops and Rpeak of 34 Tflops-31 Tflops Tflops, Infiniband - QDR(40Gb/S) {Qlogic}36 Port Federated (36*36)/2=648 with 114T (HP) Storage. |
| Operating System | RedHat Enterprise Linux (RHEL) 6.6 |
| Specific Usage | High Performance Computing with Several Softwares |
HPC2010 Cluster |
|
| Host Name | hpc2010.hpc.iitk.ac.in |
| Configuration | 368+96 Nodes Cluster with Intel Xeon 2.93Ghz 2.6Ghz; 2944/8 (non-vector) + 1536/16 (vector) Cores with Rmax of 29 TTflops ~= 25 Tflops and Rpeak of 34 Tflops-31 Tflops Tflops, Infiniband - QDR(40Gb/S) {Qlogic}36 Port Federated (36*36)/2=648 with 114T (HP) Storage. |
| Operating System | RedHat Enterprise Linux (RHEL) 6.6 |
| Specific Usage | High Performance Computing with Several Softwares |
GPU Servers |
|
| Host Name | gpu01.cc.iitk.ac.in gpu02.cc.iitk.ac.in gpu04.cc.iitk.ac.in |
| Configuration | Four Tesla T10 GPUs added to each machines with 8 cores in each |
| Operating System | Linux |
| Specific Usage | Parallel Programming under linux using CUDA with C Language |
SMP |
|
| Host Name | pawan.cc.iitk.ac.in |
| Configuration | HP Montecito 1.6 GHz/18 MB Dual Itanium Processor, 128 GB DDR2 RAM, 4x300 GBSCSI Hard Disk |
| Operating System | HPUX |
| Specific Usage | Parallel Computing with Fortran, C, C++ Compilers. |
MIC (Intel Xeon-phi-Coprocessor) |
|
| Host Name | mic.hpc.iitk.ac.in |
| Configuration | 4 Nodes Cluster with Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz and 16 cores per node; Each node contain two Intel Xeon-phi-Coprocessor cards (120 cores per card with core frequency 1.05 GHZ), Infiniband - FDR (56Gb/S) {Mellanox} 36P |
| Operating System | RedHat Enterprise Linux (RHEL) 6.6 |
| Specific Usage | Parallel Computing using MIC architecture with Intel compiler |
| How to Use | Contact to HPCGROUP,Room No.212,Computer Centre |
Parallel computer programs are more difficult to write than sequential ones, because concurrency introduces several new classes of potential software bugs, of which race conditions are the most common. Communication and synchronization between the different sub-tasks are typically some of the greatest obstacles to getting good parallel program performance.