CN102541994A - Method for deploying and sharing NFS (network file system) automatically and quickly - Google Patents

Abstract

The invention provides a method for deploying and sharing an NFS (network file system) automatically and quickly. On the basis of Shell Script, the interactive environment is provided, test parameters and a performance reference table required by the Lin Pack can be acquired simply conveniently by only inputting cluster configuration parameters according to prompts, the theoretical floating-point peak value and the actual floating-point peak value can be acquired by the Lin Pack, concurrency efficiency of the system can be further determined and the purpose of evaluating the cluster system quickly is achieved. A deployer only needs to place scripts and files required by deployment on management nodes, writes host names of all nodes (management nodes and computing nodes) into a host file, writes to-be-shared directory names into a share_dir file, runs the install.sh by a root user on the management nodes, inputs the root passwords, so that sharing and deployment of the NFS of all nodes can be completed and are still available in the next starting.

Description

The method that a kind of automated quick deployment NFS shares

Technical field

The present invention relates under the Linux cluster environment, to the deployment of NFS (NFS).Be specifically related to automated quick deployment based on the NFS (NFS) of shell script and expect automatic answer function.

Background technology

A lot of processors are used in high-performance calculation (HPC) usually, in the high speed internet network, use parallel computation environments such as MPI, operation parallel computation software, the efficient of acceleration section mathematical operations.Be accompanied by HPC the popularizing of colleges and universities and scientific research institutions, set up the large-scale cluster system of high parallel efficiency, causes such as scientific research, education, national defence are had great significance.So fast and effectively the performance of a cover group system being assessed is vital to the large-scale cluster system of setting up high parallel efficiency.

Weigh an important indicator of computing power and calculate peak value exactly, Floating-point Computation peak value for example, it is meant the Floating-point Computation maximum times that computing machine can be accomplished p.s..Comprise theoretical floating-point peak value and actual measurement floating-point peak value: theoretical floating-point peak value is can accomplish the Floating-point Computation maximum times p.s. that can reach on this computer theory; It mainly is by the decision of the dominant frequency of CPU, and each clock period of theoretical floating-point peak value=CPU frequency * CPU is carried out core cpu number in the number of times * system of floating-point operation.

Actual measurement floating-point peak value is meant the Linpack test value, that is to say operation Linpack test procedure on this machine, the test result of the optimum that obtains through the excellent method of various accent.In fact in the practical programs operational process, possibly reach actual measurement floating-point peak value hardly, say nothing of and reached theoretical floating-point peak value.These two values are just as an index weighing machine performance, are used for showing the tolerance of the scale and the potential of machine processing ability.

In the tradition LinPack test, there are some test parameters (for example most important N (problems sizes)) to need to confirm that the unfamiliar people of knowwhy is easy to calculate wrong parameter, causes whole test crash according to concrete cluster environment.Simultaneously, after testing out the actual operation peak-peak, also quite complicated according to the performance of outcome evaluation whole system.

Summary of the invention

The purpose of this invention is to provide the method that a kind of automated quick deployment NFS shares.

The objective of the invention is to realize by following mode; The deployer disposes required script and file with all and is placed on the management node; All nodes are comprised that the host name of management node and computing node writes among the hostfile, and the directory name that needs are shared: comprise home, opt is written in the share_dir file; On management node, move install.sh with root user; Behind the input root password, the NFS NFS that can accomplish at all nodes shares and maps out the work, and still effective after start next time;

Concrete job step is following:

1) all is disposed required file, comprise 6 script files and 2 configuration files, be put under the same catalogue on the management node;

2) in hostfile, write the host name of all nodes, write management node earlier, write computing node again; In share_dir, writing needs shared directory name, comprises opt, home;

3) use root user to move the install.sh file, and according to prompting input root password;

4) the install.sh script can call ssh_conf1.exp, ssh_conf2.exp, these three auto answer scripts based on expect of ssh_conf3.exp automatically, realizes the trust visit of the direct ssh of all nodes;

5) the install.sh script then can call mu_conf.exp automatically, accomplishes management node, promptly NFS_Server hold configuration task;

6) the install.sh script then can call cu_conf.exp automatically, accomplishes all computing nodes, promptly NFS_clietn hold configuration effort;

7) behind the above all working of correct completion, the install.sh script can be pointed out cluster NFS and disposed and accomplish, and this moment, system normally used the NFS function.

The invention has the beneficial effects as follows: the present invention uses the method based on Shell Script; Interactive environment is provided; Only need to obtain required test parameter of LinPack and performance reference table simply and easily, re-use Linpack and obtain theoretical floating-point peak value and actual floating-point peak value according to prompting input cluster configuration parameter; And further confirm the parallel efficiency of system, reach the purpose of rapid evaluation group system.

Summary of the invention

The NFS (NFS) that the objective of the invention is to dispose automatically, fast under the cluster environment shares.What the deployer need do only is that all are disposed required script and file (6 script files; 2 configuration files) be placed on the management node; The host name of all nodes (management node and computing node) is write among the hostfile, the directory name that needs are shared (as/home ,/opt etc.) be written in the share_dir file; On management node, move install.sh with root user; Behind the input root password, the NFS (NFS) that can accomplish at all nodes shares and maps out the work, and still effective after start next time.

Concrete job step of the present invention is following:

The first step is disposed required file (6 script files, 2 configuration files) with all and is put under the same catalogue on the management node;

Second step write the host name of all nodes in hostfile, write management node earlier, write computing node again; In share_dir, writing needs shared directory name, as/opt ,/home etc.;

In the 3rd step, use root user to move the install.sh file.And according to prompting input root password;

In the 4th step, the install.sh script can call ssh_conf1.exp, ssh_conf2.exp, these three auto answer scripts based on expect of ssh_conf3.exp automatically, realizes the trust visit of the direct ssh of all nodes;

In the 5th step, the install.sh script then can call mu_conf.exp automatically, accomplishes management node, promptly NFS_Server hold configuration task;

The 6th the step, the install.sh script then can call cu_conf.exp automatically, accomplishes all computing nodes, promptly NFS_clietn hold configuration effort;

In the 7th step, behind the above all working of correct completion, the install.sh script can be pointed out cluster NFS and disposed and accomplish, and can normally use the NFS function this moment.

Description of drawings

Fig. 1 is a network architecture diagram;

Fig. 2 disposes process flow diagram.

Embodiment

Hardware environment: management node: 1

Computing node: 8

Every node memory 8GB

Every node check figure: 12 cores

Cpu frequency: 2.66GHz

Network connects: Infiniband QDR

In hostfile, write management node host name and whole computing node host name, as shown in the table:

[rootmu01?NFS_Express]#?cat?hostfile

mu01

cu01

cu02

cu03

cu04

cu05

cu06

cu07

cu08

In share_dir, writing the catalogue of need sharing, this example shares/opt and/home, operation steps is following:

[rootmu01?NFS_Express]#./install.sh

Installation?is?Starting……

Please?gimme?root’s?password?:

111111

Conform?root’s?password?；

111111

Now，Trusted_SSH?configuration?is?in?procession,Please?What….

send—“111111”\r””

send:spawn?id?exp6?not?open

while?executing

“send?–111111\r”

******Trusted_SSH?Configuration?is?COMPLETE*****

************************************************

Now,NFS_Server?configuration?is?in?procession?,Please?Wait…

************************************************

****NFS_Server?Configuration?is?COMPLETE***

************************************************

Now,NFS_Clinents?Configuration?is?in?procession,Please?Wait…

rc.local

rc.local

rc.local

************************************************

***?NFS_Clinents?Configaration?is?COMPLETE***

************************************************

Starting?NFS?services:

Starting?NFS?quotas:

Starting?NFS?daemon:

Starting?NFS?mountd:

Starting?portmap:

Starting?portmap:

************************************************

***?NFS?Configaration?is?COMPLETE***

************************************************

On management node, move install.sh with the root identity, the root password is 111111, accomplishes configuration effort.

When seeing NFS Configuration is COMPLETE prompting, configuration all completion are described.Check that to computing node all computing nodes are carry assigned catalogue all, and be saved in the automatic carry of start.Be described below:

[rootcu01～]#?mount

/dev/mapper/VolGroup00-L0go100?on?/type?ext3(rw)

Proc?on?/proc?type?sysfs?(rw)

Devpts?on?/sys?type?devpts(rw,gid=5,mode=620)

/dev/hda1?on?/doot?type?ext3(rw)

Tmpfs?on?/dev/shm?type?tmpfs(rw)

none?on?/proc/sys/fs/binfmt_misc?type?binfmt_misc(rw)

sunrpc?on?/var/lib/nfs/rpc_pipefs?type?rpc_pipefs(rw)

mu01:/opt?on?/home?type?nfs(rw,addr=192.168.1.105)

mu01:/home?on?/home?type?nfs?(rw,addr=192.168.1.105)

The automatic carry option of starting shooting is opened:

[rootcu01～]#?cat?/etc/rc.local

#!/bin/sh

#

#?This?script?will?be?executed?*after*?all?the?other?init?scripts.

#?You?can?put?your?own?initialization?stuff?in?here?if?you?don’t

#?want?to?do?the?full?sys?V?style?init?stuff?.

Touch?/var?/lock?/subsys/local

Mount?–t?nfs?mu01:/opt/opt

Mout?–t?nfs?mu01:/home/home

So far, accomplished whole configurations of NFS, can normally use.And all configuration take-effectives when guaranteeing the next time start.

Except that the described technical characterictic of instructions, be the known technology of those skilled in the art.

Embodiment

Explanation at length below with reference to Figure of description method of the present invention being done.

The method that automated quick deployment NFS of the present invention shares, it is 2 parts that the network architecture is divided into: storage networking and computational grid;

Computational grid adopts the Infiniband switch, the HCA through IB cable collocation server link receive computing node (d1---dn).Adopt IB over IP communication mechanism, realize internodal express network communication.

Storage networking uses 8GB FC optical fiber switch, adopts the memory device of FC SAN framework, divides a plurality of LUN, maps directly to management node and uses.

Claims (1)

1. the method shared of an automated quick deployment NFS; It is characterized in that the deployer disposes required script and file with all and is placed on the management node; All nodes are comprised that the host name of management node and computing node writes among the hostfile, and the directory name that needs are shared: comprise home, opt is written in the share_dir file; On management node, move install.sh with root user; Behind the input root password, the NFS NFS that can accomplish at all nodes shares and maps out the work, and still effective after start next time;

Concrete job step is following:

1) all is disposed required file, comprise 6 script files and 2 configuration files, be put under the same catalogue on the management node;

2) in hostfile, write the host name of all nodes, write management node earlier, write computing node again; In share_dir, writing needs shared directory name, comprises opt, home;

3) use root user to move the install.sh file, and according to prompting input root password;

4) the install.sh script can call ssh_conf1.exp, ssh_conf2.exp, these three auto answer scripts based on expect of ssh_conf3.exp automatically, realizes the trust visit of the direct ssh of all nodes;

5) the install.sh script then can call mu_conf.exp automatically, accomplishes management node, promptly NFS_Server hold configuration task;

6) the install.sh script then can call cu_conf.exp automatically, accomplishes all computing nodes, promptly NFS_clietn hold configuration effort;

7) behind the above all working of correct completion, the install.sh script can be pointed out cluster NFS and disposed and accomplish, and this moment, system normally used the NFS function.

Images