HIGH PERFORMANCE COMPUTING CLUSTER

HPC Cluster is a group of computing nodes joins computational powers of them to provide a more combined computational power. Therefore, cluster computing utilizes multiple machines to provide a more powerful computing environment perhaps through a single operating system. In its simplest structure, as said above the HPC clusters are intended to utilize parallel computing to apply more processor force for the solving the problem. Though from the outside the cluster may look like a single system, the internal workings to make this happen can be quite complex. The idea is that the individual tasks that make up a parallel application should run equally well on whatever node they are dispatched on. However, some nodes in a cluster often have some physical and logical differences. High-Performance cluster at Central computer center has a 14 node architecture; One master node, 6 GPU Compute node, 6 compute nodes, 1 Xeon Phi (KNL) node.


hpc
HPC
hpc
Diagram

Specifications

Master /Head Node: [Quantity 1]

Processor

2X Intel Xeon Processor (E5-2640 v3 8C 2.6GHz,20MB 1866MHz 90W)

Chipset

Intel C610 Chipset

Accelerator

1x NVIDIA Tesla K20 GPU or higher

Memory

x 8GB TruDDR4 Memory (1Rx4, 1.2V) PC4-17000 CL15 2133MHz LP RDIMM

Internal Storage

8x 1TB 7.2K 6Gbps NL SATA 3.5" G2HS HDD

PCI-E slot

Support for minimum 2 Nos of NVIDIA Kepler GPU cards (2x PCI-E 3.0 x16 slots)

Interfaces

2 x USB 3.0 (rear), 1 x VGA (1x rear), 3 / 2x Gbit/s Ethernet + 1x service LAN Onboard, Management LAN traffic can be switched to shared onboard Gbit LAN port

Power Supply

Redundant Platinum Level (94%) Power Supplies

Form Factor

1U/2U Rack Server with rack mount rail kit


CPU Compute Node: [Quantity 5]

Processor

Dual 64 Bit Processor from x86 family, Intel Broadwell E5-2640v4

Memory

8x8GB TruDDR 4, 2133MHz scalable upto 1.5TB

Internal Storage

500GB, SATA disk scalable up to 8 disk

Interfaces

FDR port with 3 Mtr copper cable

Operating System

64-bit CentOS 6.7

System Management

Advanced failure analysis support on systems for CPU, memory, HDD, Power supply and fans.


GPU Compute Nodes: [Quantity 6]

Processor

2X Intel Xeon Processor E5-2640 v3 8C 2.6GHz 20MB 1866MHz 90W

Accelerator

2x NVIDIA K20c, 5GB, 2496 Cores

Memory

8x 8GB TruDDR4 Memory (1Rx4, 1.2V) PC4-17000 CL152133MHz LP RDIMM

Internal Storage

1x 1TB 7.2K 6Gbps NL SATA 3.5" G2HS HDD

Network

Broadcom Corporation NetXtreme BCM5719 Gigabit Ethernet PCIe

Advanced Interface

Mellanox Technologies MT27520 Family (FDR)

Form Factor

1 x Intel Xeon Phi 7250 Processor(272 cores)

OS Support

Fully certified/compatible with the latest RHEL 7. x

Memory

64 GB(16GB x 4) DDR4, 2400 MHz ECC Memory

Internal Storage

480 GB

Network

Two 1GbE network ports with PXE boot capability

Form Factor

Half-width 1U or equivalent (e.g. 4 servers in 2U) rack-mountable


Storage: [Quantity 1]

Processor

2X Intel Xeon Processor

Memory

2x 4GB PC3L- 10600E

Internal Storage

16X 1 TB 7,200 rpm 6 Gb SAS NL 2.5 Inch HDD (12 TB Useable after RAID 5 Fiber channel storage)

Interface

8Gb FC 4 Port Host Interface Card with FC


Infiniband Switch

Model-Mellanox

SX6036

36 FDR (56Gb/s) ports in a 1U switch

4.032Tb/s switching capacity

FDR/FDR10 support for Forward Error Correction(FEC)

Remote Management Tool CLI, SNMP

Port Mirroring


Ethernet Switch

Model

Lenovo RackSwitch G7028

Performance

128 Gbps switching throughput (full duplex) Latency of 3.3 microseconds 96 Mpps

Interface Options

24 × 1 GbE (24 RJ-45), 4 × 10 GbE SFP+