commit | a19ff2508e37fce5e4909e0b717c3a4d2f0be679 | [log] [tgz] |
---|---|---|
author | Joel Fernandes <joelaf@google.com> | Mon May 15 03:02:31 2017 -0700 |
committer | KP Singh <kpsingh@google.com> | Thu Jun 29 22:51:02 2017 +0200 |
tree | 19790e61fcd2b070621c4780bf5c8c0b0c44c026 | |
parent | 50610cbe8742f4a1e3da60b406cb36b4c962bd72 [diff] |
trappy: Speed up trappy by caching trace parsing Pandas is extremely fast at parsing csv to data frames. Astonishingly it takes < 1s to serialize/deserialize a 100MB work of traces with 430000 events to/from csv. We leverage this and write out a data frames into a csv file when they are created for the first time. Next time we read it out if it exists. To make sure, the cache isn't stale, we take the md5sum of the trace file and also ensure all CSVs exist before reading from the cache. I get a speed up of 16s to 1s when parsing a 100MB trace. Co-developed-by: Brendan Jackman <brendan.jackman@arm.com> Signed-off-by: Joel Fernandes <joelaf@google.com> Reviewed-by: KP Singh <kpsingh@google.com>
TRAPpy (Trace Analysis and Plotting in Python) is a visualization tool to help analyze data generated on a device. It parses ftrace-like logs and creates in-memory data structures to be used for plotting and data analysis.
The following instructions are for Ubuntu 14.04 LTS but they should also work with Debian jessie. Older versions of Ubuntu or Debian (e.g. Ubuntu 12.04 or Debian wheezy) will likely require to install more packages from pip as the ones present in Ubuntu 12.04 or Debian wheezy will probably be too old.
$ sudo apt install trace-cmd kernelshark
$ sudo apt install python-pip python-dev
$ sudo apt install libfreetype6-dev libpng12-dev python-nose $ sudo pip install numpy matplotlib pandas ipython[all]
$ sudo pip install --upgrade trappy
Now launch the ipython notebook server:
$ ipython notebook
This should pop up a browser. If it doesn't, open a web browser and go to http://localhost:8888/tree/
In the doc/
folder there's a 00 - Quick start
which describes how to run TRAPpy. Other notebooks in that directory describe other functions of TRAPpy.
API reference can be found in https://pythonhosted.org/TRAPpy/
The code of the TRAPpy toolkit with all the supported tests and Notebooks can be cloned from the official GitHub repository with this command:
$ git clone https://github.com/ARM-software/trappy.git
An easy way to test your installation is to use the nosetests
command from TRAPpy's home directory:
$ nosetests
If the installation is correct all tests will succeed.