Shulou(Shulou.com)06/02 Report--

Written on June 13, 2017

Author: Zhai Jiangheng

In general, HPC will not be installed manually, but hpc will be deployed with integrated automation solutions, but these integration solutions are integrated with specified operating system distributions, such as the common HPC integrated automation solution rockets, so far integrated centos 6.6.If you need to install HPC in other linux distribution environments, you cannot use integrated automation solutions, you must use manual installation methods This blog is about how to build HPC manually. Please read on.

1. Brief introduction to HPC

2. HPC architecture

3. Ssh password-free access

4. Install and configure NFS

5. Install and configure NIS

6. Install tentakel

7. Install openmpi

8. Install torque

9. Install ganglia

1. Brief introduction to HPC

Highperformance computing (HPC) refers to a computing system and environment that usually uses many processors (as part of a single machine) or several computers organized in a cluster (operating as a single computing resource). There are many types of HPC systems, ranging from large clusters of standard computers to highly specialized hardware. Most cluster-based HPC systems use high-performance network interconnections, such as those from InfiniBand or Myrinet. Basic network topologies and organizations can use a simple bus topology. In high-performance environments, mesh network systems provide short latency between hosts, so overall network performance and transmission rates can be improved.

2. Hpc architecture

The HPC network architecture is shown in the figure

The figure above shows a mesh HPC system. In a mesh network topology, this structure supports faster communication across hosts by shortening the physical and logical distance between network nodes.

Although network topology, hardware, and processing hardware are important in HPC systems, the core functions that make the system so effective are provided by the operating system and application software.

HPC systems use specialized operating systems that are designed to look like individual computing resources. As you can see from figures 1 and 2, there is a control node that forms the interface between the HPC system and the client. The control node also manages the work distribution of the computing node.

For task execution in a typical HPC environment, there are two models: single instruction / multiple data (SIMD) and multiple instruction / multiple data (MIMD). SIMD performs the same evaluation instructions and operations simultaneously across multiple processors, but for different data ranges, it allows the system to evaluate the same expression using many variables at the same time. MIMD allows the HPC system to perform different calculations using different variables at the same time, making it look like the whole system is not just a computing resource without any features (although it is powerful), and can perform many calculations at the same time.

Whether using SIMD or MIMD, the basic principle of a typical HPC is still the same: the entire HPC unit operates and behaves like a single computing resource, expanding the actual requested load to each node. HPC solutions are also dedicated units that are specifically designed and deployed to act as (and only serve as) large computing resources.

3.SSH password-less access

Set up a script directory

Mkdir-p / root/rootUser

Mkdir-p / home/ ordinary user directory

For example: mkdir-p / home/hpc (hpc is an ordinary user)

Write the following three scripts:

Setp1.exp

The contents are as follows:

#! / bin/bash

Rm-fr/root/.ssh

/ usr/bin/expect

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.