+5 votes
292 views
How to install and use Guider (Performance Analyzer) on Linux

in Linux / Unix by (551k points)
reopened | 292 views

1 Answer

+3 votes
Best answer

1. Install Guider on Linux
2. Configure the Guider buffer in Linux
3. Perform CPU usage analysis in thread mode in Linux
4. Perform real-time Linux monitoring in Top Linux mode
5. Analyze a simple Guider process using PID in Linux
5. Save Guider results to a file in Linux
6. Ways to use Guider on Linux
7. General Guider options on Linux

Maintaining a constant analysis of the system and users that we manage will undoubtedly be one of the best practices that we can perform as administrators because we have the opportunity to know in real time the behavior and state of the system and each of its components..

In this area we have the option of using the Guider tool thanks to which the administration and performance analysis options will be much more complete and we will have the possibility to know in detail if any component has any animalia to act before failures occur .

In this tutorial we will see how to install and use Guider in Linux environments..

What is Guider?
Guider is a Linux open source performance analyzer that was developed to measure the use of system resources and, taking these analyzes, give users advice to improve the performance of the operating system used.

Guider is able to track and analyze the use of thread, process and function resources in a detailed and complete way.

Guider is written in Python for Linux operating systems..

Guider features
Among the features that stand out in Guider we have:
  • Easy to use: Its use is simple, just run the utility without installing or configuring anything additional
  • Correct measurements: With Guider it is possible to measure the time in ms and the size in MB / KB
  • Integrated functions: Using Guider it will be possible to deploy the use of CPU, memory or disk by thread, process or function (user / kernel).
Requirements to use Guider
Before using Guider we must have the following requirements:
  • Linux kernel (> = 3.0)
  • Python (> = 2.7)
  • Kernel buffer size = 40960.

To know the version of Python in our system, in this case Ubuntu 17 Server, we will execute the following:

 python3 –version 
To know the version of Kernel we execute:
 join me -r 

image

In case of not having Python we can install it by executing the following command:

  • Ubuntu
 sudo apt-get install python3.4 
  • Fedora
 sudo dnf install python3.4 
  • OpenSUS
 sudo zypper install python3 

1. Install Guider on Linux

Step 1

To install Guider correctly it will be necessary to clone the Guider repository located on GitHub by running the following commands:
 git clone https://github.com/iipeace/guider.git 

image

Step 2

Then we will access the directory of Guider by running:
 cd guider 
Step 3

Once there we have the following options:
  • If we want to run Guider without installation we must execute the following:
 guider.py 
  • If we want to install it we proceed to execute the following:
 make sudo make install 
  • If we use PIP in the distribution we can execute the following:
 sudo pip install --pre guider 
Step 4

For this example we will use the guide.py option to not install the utility on the system, for its startup we will execute the following:
 sudo /guider.py 

2. Configure the Guider buffer in Linux

By default, guider must configure the buffer size for the operations performed on the system, however, in case this does not happen, an error will be generated at the time of execution.

Step 1

We can verify the size of the buffer with the following command:
 sudo cat / sys / kernel / debug / tracing / buffer_size_kb 

image

Step 2

In case the value is less than 40960 we must execute the following line to establish this value:
 echo 40960 | sudo tee / sys / kernel / debug / tracing / buffer_size_kb 
image
Step 3

With this defined we can start Guider using the line mentioned above:
 sudo /guider.py 
image
Step 4

It will be possible to use guider in the thread, function, startup, file and system modes using the following syntax:
 guider [mode | file] [options] 

3. Perform CPU usage analysis in thread mode in Linux

Step 1

With the following command we will start the sequence tracking process which we can end by pressing the Ctrl + C keys, after this, the data will be saved and the analysis process will begin to finally display the analysis report.
 sudo guider record 
image
Step 2

By pressing Ctrl + C the data will be stored and we will automatically see the results: image
Step 3

We can see details such as date, kernel, system load time and many more parameters. We can go forward in the report with the AvPag keys or go back with RePag: image
Step 4

As we move forward we will see new details of the system used. image

4. Perform real-time Linux monitoring in Top Linux mode

Step 1

With the following command we can see the use of Linux process resources in real time:
 sudo guider.py top 
image
Step 2

It will be possible to set an interval to show the output using the -i switch followed by the seconds:
 sudo guider top -i 4 

5. Analyze a simple Guider process using PID in Linux

Step 1

To perform this process, we must first know the process PID, in this Apache example, for this we can execute any of the following options:
 pidof apache2 ps -e | grep apache2 
image
Step 2

We will see the list of associated processes, now, we can use the following line which generates the CPU cycle, the instruction number, the IPC, the failures, the lack of cache, the real-time branch failure:
 sudo guider top -eP -g 1943 
Step 3

The -g parameter filters by the indicated process. image

5. Save Guider results to a file in Linux


For administration tasks, it is possible to save the tracking data or any output in a file for its respective analysis.
Step 1

With the following command, we will save the trace data in a file called guider.dat, defined by default, in the current directory, but if we wish we can set a different path:
 sudo guider -s. 
Step 2

To save another output in a file called guider.out, by default, in the current directory we will use the following line:
 sudo guider top -o. 
Step 3

At the moment we want to see the contents of these files we can execute the following commands:
 cat guider.dat cat guider.out 
Step 4

In many it will be possible that we want to get the help of Guider, this will be obtained with the following command:
 sudo ./guider.py -h 
image

6. Ways to use Guider on Linux


Guider allows us to use various operating modes such as:
 top [real time] record [thread] record -and [system] record -f [function] record -F [file] view [page] 

7. General Guider options on Linux

-and
Enable options according to mode.
-d
Disable options according to mode.
-s
Save data
-S
Sort the data output with values ​​such as .c (pu) / m (em) / b (lock) / w (fc) / p (id) / n (ew) / r (untime)

-

-or
It runs in the background.
-W
Waiting for a signal.
-R
Record continuously.
-b
Set the buffer size.
-D
It generates a dependency on a thread.
-t
Analyze the Syscall values.
-T
Set the source path.
-j
Define the report path.
-OR
Generate user event values.
-K
Generate kernel event values.
-x
Define the address of the local server.
-P
Group by processes.
-q
Configure task list.
-L
Convert text to images.
-to
Show all information.
-i
Set intervals in seconds.
-g
Add filters
-C
Define custom events.

Thus, Guider is an additional help to the common administration tasks offering us possibilities to obtain global or real-time results of the elements of the system and thus manage them in a much more direct way.


by (3.5m points)
edited

Related questions

+4 votes
1 answer
+5 votes
1 answer
asked Apr 30, 2020 in Linux / Unix by backtothefuture (551k points) | 294 views
+3 votes
1 answer
+4 votes
1 answer
asked Nov 11, 2019 in Linux / Unix by backtothefuture (551k points) | 365 views
+3 votes
1 answer
Sponsored articles cost $40 per post. You can contact us via Feedback
10,634 questions
10,766 answers
510 comments
3 users