Software Open Access

bio-phys/MDBenchmark: Release 2.0.0

Michael Gecht; Max Linke; Marc Siggel

JSON-LD ( Export

  "description": "<p>After lots of work, MDBenchmark 2.0 is finished. This new release includes major improvements to the plotting functionality and user experience in general. But it also comes with breaking changes to the command-line interface and requires an update to your Python environment.</p>\n<p>If you have any workflows that rely on the function of specific calls to <code>mdbenchmark</code>, please check the <a href=\"\">latest CLI reference</a> to make sure everything works as expected.</p>\nHighlights\n<ul>\n<li>Revamped plotting functionality, making it possible to plot many different benchmark combinations (modules, hosts, other settings) at once.<ul>\n<li>The old <code>mdbenchmark analyze --plot</code> command was replaced with <code>mdbenchmark plot</code>.</li>\n<li>Plots can now be easily customized with the <code>--dpi</code>, <code>--font-size</code>, <code>--xtick-step</code>, <code>--no-fit</code> and <code>--no-watermark</code> options.</li>\n</ul>\n</li>\n<li>All calls to the main <code>mdbenchmark</code> commands (<code>generate</code>, <code>analyze</code>, <code>submit</code> and <code>plot</code>) now use consistent option names, when handling similar functions.</li>\n<li>Running <code>mdbenchmark analyze</code> will not produce CSV output files, until asked to with the <code>--save-csv</code> option.</li>\n<li><code>mdbenchmark generate</code> and <code>mdbenchmark submit</code> will now ask for confirmation, before performing any operations. The confirmation can be skipped with the <code>--yes</code> option.</li>\n<li>CPU and GPU benchmarks can now be generated at once, without the need to run two separate commands: <code>mdbenchmark generate --cpu --gpu</code>.</li>\n<li>The job name for the queuing system can be customized with the <code>--job-name</code> option when using <code>mdbenchmark generate</code>.</li>\n</ul>\nInstallation\n<p>Please refer to <a href=\"\">our documentation</a> on how to install the package from your favorite package manager.</p>\nMigrating\n<p>When upgrading from a previous version, you need to remove three packages that were previously installed alongside of MDBenchmark: <code>datreant.core</code>, <code></code> and <code>mdsynthesis</code>. Refer to our <a href=\"\">upgrading guide</a>, if you need assistance.</p>\nUpgrading\n<p>To upgrade, use the appropriate command for your Python package manager.</p>\n<code>conda</code>\n<p>If you are using <code>conda</code> and your conda environment is called <code>benchmark</code>, simply run:</p>\n<pre><code class=\"lang-bash\">conda update -n benchmark mdbenchmark\n</code></pre>\n<code>pip</code>\n<p>Users of <code>pip</code> run the following:</p>\n<pre><code class=\"lang-bash\">pip install mdbenchmark --upgrade\n</code></pre>\nFull changelog\n<p>A full list of changes, both user facing and behind the scenes, can be found below.</p>\nFeatures\n<ul>\n<li><p>Updated plotting interface. <code>mdbenchmark analyze --plot</code> is not deprecated. Use <code>mdbenchmark plot</code> instead.</p>\n<p>The new workflow for plotting data is as follows:</p>\n<p>1) Use <code>mdbenchmark analyze --save-csv results.csv</code> to generate a CSV output file.\n2) Use <code>mdbenchmark plot --csv results.csv</code> to plot the data.</p>\n<p>Consult <code>mdbenchmark &lt;command&gt; --help</code> for options to filter your data accordingly. (#52)</p>\n</li>\n<li><code>mdbenchmark generate</code> now accepts <code>--cpu</code> / <code>--no-cpu</code> and <code>--gpu</code> / <code>--no-gpu</code>. The default is <code>--cpu</code> and <code>--no-gpu</code>. (#69)</li>\n<li>Added user prompts to <code>mdbenchmark generate</code> and <code>mdbenchmark submit</code>. (#90)</li>\n<li>Added <code>--yes</code> flag to <code>mdbenchmark generate</code> and <code>mdbenchmark submit</code> to bypass user prompt. (#90)</li>\n<li>Added <code>-nc</code> and <code>-ng</code> options to <code>mdbenchmark generate</code>. These are short hand for <code>--no-cpu</code> and <code>--no-gpu</code>, respectively. (#93)</li>\n<li>Added template for MPCDF cluster <code>cobra</code>. (#104)</li>\n<li>Added <code>--template</code> and <code>-t</code> option to <code>mdbenchmark generate</code>, to specify a job template. The <code>--host</code> option still works. (#106)</li>\n<li>Standarize the CLI options across all <code>mdbenchmark</code> calls. (#107)</li>\n<li>Added <code>mdbenchmark plot --dpi</code> option to change the plot DPI. (#108)</li>\n<li>Added <code>mdbenchmark plot --font-size</code> to change the plot font size. (#108)</li>\n<li>Linear scaling fit can now be hidden with <code>--no-fit</code>. (#108)</li>\n<li>Updated <code>ylim</code>, <code>xtick</code>  and <code>ytick</code> defaults. The steps for <code>xtick</code> can be overwritten with <code>mdbenchmark plot --xtick-step</code>. (#108)</li>\n<li>Added a watermark in the top left corner for every plot. Can be easiliy disabled with <code>mdbenchmark plot --no-watermark</code>. (#108)</li>\n<li><code>mdbenchmark analyze</code> no longer writes CSV files by default. <code>--save-csv</code> flag added to write csv files. (#119)</li>\n<li>Added <code>mdbenchmark generate --job-name</code> to change the job name submitted to the queuing system. (#125)</li>\n</ul>\nBugfixes\n<ul>\n<li>Fixed a bug where benchmark creation with files ending in <code>.namd</code> did not work. (#124)</li>\n<li>Fixed a bug where benchmark creation would fail when the input file was not in the current directory. (#124)</li>\n</ul>\nMisc\n<ul>\n<li>Replaced <code>mdsynthesis</code> with <code>datreant</code> and upgraded to the new <code>datreant&gt;=1.0</code> format. (#110)</li>\n</ul>", 
  "license": "", 
  "creator": [
      "affiliation": "MPI of Biophysics", 
      "@type": "Person", 
      "name": "Michael Gecht"
      "affiliation": "", 
      "@type": "Person", 
      "name": "Max Linke"
      "affiliation": "Max Planck Institute of Biophysics", 
      "@type": "Person", 
      "name": "Marc Siggel"
  "url": "", 
  "codeRepository": "", 
  "datePublished": "2018-10-30", 
  "version": "v2.0.0", 
  "@context": "", 
  "identifier": "", 
  "@id": "", 
  "@type": "SoftwareSourceCode", 
  "name": "bio-phys/MDBenchmark: Release 2.0.0"
All versions This version
Views 340158
Downloads 462
Data volume 9.6 MB357.5 kB
Unique views 282124
Unique downloads 152


Cite as