Skip to content

Commit d48ba5f

Browse files
author
Elias Werner
committed
add a bit of readme
1 parent 7b6ea75 commit d48ba5f

1 file changed

Lines changed: 36 additions & 11 deletions

File tree

README.md

Lines changed: 36 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -6,12 +6,19 @@
66
<img width="450" src="doc/JUmPER01.png"/>
77
</p>
88

9-
# A Jupyter Kernel for Performance Engineering
9+
# JUmPER: Jupyter meets Performance
1010

11-
This is the JUmPER Kernel that enables you to instrument and trace or profile Jupyter cells with [Score-P](https://score-p.org/).
11+
JUmPER brings performance engineering to Jupyter. It consists of the two repositories:
1212

13-
For binding to Score-P, the kernel uses the [Score-P Python bindings](https://github.com/score-p/scorep_binding_python).
13+
- JUmPER Ipython extension (https://github.com/ScaDS/jumper_ipython_extension)
14+
15+
The extension allows you to take coarse-grained performance measurements and create visualizations in your Python kernels.
16+
17+
- Score-P Jupyter kernel Python (this repository)
1418

19+
The Score-P kernel allows you to instrument, and trace or profile your Python code in Jupyter using [Score-P](https://score-p.org/).
20+
21+
For binding to Score-P, the kernel uses the [Score-P Python bindings](https://github.com/score-p/scorep_binding_python).
1522

1623

1724
# Table of Content
@@ -39,8 +46,8 @@ For binding to Score-P, the kernel uses the [Score-P Python bindings](https://gi
3946
To install the kernel and required dependencies:
4047

4148
```
42-
pip install jumper-kernel
43-
python -m jumper.install
49+
pip install scorep_jupyter
50+
python -m scorep_jupyter.install
4451
```
4552

4653
You can also build the kernel from source via:
@@ -50,20 +57,29 @@ pip install .
5057
```
5158

5259
The kernel will then be installed in your active python environment.
53-
You can select the kernel in Jupyter as `jumper`.
60+
You can select the kernel in Jupyter as `Score-P_Python`.
5461

5562
**For using the Score-P features of the kernel you need a proper Score-P installation.**
5663

57-
```
58-
pip install scorep
59-
```
60-
6164
From the Score-P Python bindings:
6265

6366
> You need at least Score-P 5.0, build with `--enable-shared` and the gcc compiler plugin.
6467
> Please make sure that `scorep-config` is in your `PATH` variable.
6568
> For Ubuntu LTS systems there is a non-official ppa of Score-P available: https://launchpad.net/~andreasgocht/+archive/ubuntu/scorep .
6669
70+
To use the coarse-grained performance measurements, simply install the JUmPER extension via:
71+
72+
```
73+
pip install jumper_extension
74+
```
75+
76+
Then load it in a Jupyter cell with:
77+
78+
```
79+
%load_ext jumper_extension
80+
```
81+
82+
You can find usage information in the repository of the extension.
6783

6884
# Usage
6985

@@ -200,7 +216,16 @@ PRs are welcome.
200216

201217
# Citing
202218

203-
If you publish some work using the kernel, we would appreciate if you cite the following paper:
219+
If you publish some work using the kernel, we would appreciate if you cite one of the following papers:
220+
221+
```
222+
Werner, E., Rygin, A., Gocht-Zech, A., Döbel, S., & Lieber, M. (2024, November).
223+
JUmPER: Performance Data Monitoring, Instrumentation and Visualization for Jupyter Notebooks.
224+
In SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 2003-2011). IEEE.
225+
https://www.doi.org/10.1109/SCW63240.2024.00250
226+
```
227+
228+
or
204229

205230
```
206231
Werner, E., Manjunath, L., Frenzel, J., & Torge, S. (2021, October).

0 commit comments

Comments
 (0)