More appealing API for use in Juypter.

notebook.ipynb

@@ -17,9 +17,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "metadata": {
-    "collapsed": true
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "import logging\n",
@@ -32,6 +30,7 @@
     "from xlap.analyse.regress import linear as linear_regression\n",
     "from xlap.analyse.trace import traces\n",
     "from xlap.analyse.correlation import correlation\n",
+    "from xlap.analyse.latency import analyse\n",
     "import pandas as pd\n",
     "%matplotlib inline"
    ]
@@ -51,7 +50,7 @@
    "source": [
     "config = parse_config()\n",
     "data_files = config[\"data_files\"]\n",
-    "original = evaluate(data_files[\"sender\"], data_files[\"receiver\"], config=config, kind=0)[:4094]"
+    "original = evaluate(data_files[\"sender\"], data_files[\"receiver\"], config=config, kind=0)"
    ]
   },
   {
@@ -103,6 +102,56 @@
     "correlation(df[df[\"EndToEnd_D\"] < 500], config)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "## Latency Criticality"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "d = analyse(original, config)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Correlations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "d.corr.sort_values(ascending=False)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Control Flow Graph"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "d.cfg"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

xlap/analyse/latency.py

 import pandas as pd
 import matplotlib.pyplot as plt
-from xlap.analyse.util import get_outlier_threshold, extract_durations, box
 from scipy.stats.stats import pearsonr
 import numpy as np
 import sys
 import graphviz
 
 
+class LatencyAnalysis():
+    def __init__(self, cfg=None, hdb=None):
+        self.cfg = cfg
+
+        correlations = []
+        labels = []
+        for x in hdb:
+            correlations += [x["Correlation"]]
+            labels += ["{} -> {}".format(x["Start"], x["End"])]
+
+        corr = pd.Series(correlations, index=labels)
+
+        self.corr = corr
+
+
 def _get_thread_for_event(config, e):
     name = str(e)[:-2]
     try:
@@ -112,13 +126,13 @@ def _plot_controlflow_graph(df, hdb):
     generate the control flow graph using dot
     """
     t_columns = [x for x in df.columns if x.endswith("_T")]
-    d = graphviz.Digraph(filename="graph", format="pdf")
+    graph = graphviz.Digraph(filename="graph", format="pdf")
     for event1 in df[t_columns]:
-        d.node(str(event1)[:-2])
+        graph.node(str(event1)[:-2])
     for edge in hdb:
-        d.edge(edge["Start"][:-2], edge["End"][:-2])
-    d.render()  # saves to graph.pdf in local folder
-    return d
+        graph.edge(edge["Start"][:-2], edge["End"][:-2])
+    graph.render()  # saves to graph.pdf in local folder
+    return graph
 
 
 # Taken from: http://composition.al/blog/2015/11/29/a-better-way-to-add-labels-to-bar-charts-with-matplotlib/
@@ -147,58 +161,53 @@ def autolabel(rects, ax, labels):
         ax.text(label_position, rect.get_y(), labels[i], ha=align, va='bottom', rotation=0, color=color)
 
 
-def _plot_critical_regions(df, hdb):
+def _plot_critical_regions(hdb):
     """
     plot regions, sorted by latency criticality
     """
     plt.rcParams["font.family"] = "serif"
-    for region in sorted(hdb, key=lambda x: -x['Correlation']):
-        print("%-10f %10s -> %10s" % (region['Correlation'], region['Start'], region['End']), file=sys.stderr)
-
-        relevant = sorted([x for x in hdb if x['Correlation'] > 0], key=lambda x: -x['Correlation'], reverse=True)
-        x = np.arange(len(relevant))
-
-        correlations = list(map(lambda x: x['Correlation'], relevant))
-        ticks = list(map(lambda x: "%s-%s" % (x['Start'][:-2], x['End'][:-2]), relevant))
-        fig, ax = plt.subplots()
-        rects = ax.barh(x, correlations, align="center", tick_label="")
-        autolabel(rects, ax, ticks)
-
-        plt.tight_layout()
-        plt.savefig("latency-criticality.pdf")
-        plt.close()
-
-    def analyse(df, config):
-        hb = []
-
-        events = [column for column in df.columns if column.endswith("_T")]
-
-        for event1 in df[events]:
-            for event2 in df[events]:
-                if str(event1) == str(event2):
-                    continue
-                if _happens_before(df, event1, event2, config):
-                    hb += [{'Start': str(event1), 'End': str(event2)}]
-
-        hdb = []
-        e2e = list(df['EndToEnd_D'])
-        for event1 in df[events]:
-            for event2 in df[events]:
-                if str(event1) == str(event2):
-                    continue
-                # if _locally_happens_directly_before(df, event1, event2, hb, config):
-                if _happens_directly_before(df, event1, event2, hb):
-                    # compute the correlation between e2e latency and event1-event2 latency
-                    l3 = list(df[event2] - df[event1])
-                    if any(map(lambda x: x != 0, l3)):
-                        correlation = pearsonr(l3, e2e)[0]
-                    else:
-                        correlation = 0
-
-                    hdb += [{'Start': str(event1), 'End': str(event2), 'Correlation': correlation}]
-
-        cfg = _plot_controlflow_graph(df, hdb)
-        _plot_critical_regions(df, hdb)
-        return {
-            "cfg": cfg
-        }
+    relevant = sorted([x for x in hdb if x['Correlation'] > 0], key=lambda x: -x['Correlation'], reverse=True)
+    x = np.arange(len(relevant))
+
+    correlations = list(map(lambda x: x['Correlation'], relevant))
+    ticks = list(map(lambda x: "%s-%s" % (x['Start'][:-2], x['End'][:-2]), relevant))
+    fig, ax = plt.subplots()
+    rects = ax.barh(x, correlations, align="center", tick_label="")
+    autolabel(rects, ax, ticks)
+
+    plt.tight_layout()
+    plt.savefig("latency-criticality.pdf")
+    plt.close()
+
+def analyse(df, config):
+    hb = []
+
+    events = [column for column in df.columns if column.endswith("_T")]
+
+    for event1 in df[events]:
+        for event2 in df[events]:
+            if str(event1) == str(event2):
+                continue
+            if _happens_before(df, event1, event2, config):
+                hb += [{'Start': str(event1), 'End': str(event2)}]
+
+    hdb = []
+    e2e = list(df['EndToEnd_D'])
+    for event1 in df[events]:
+        for event2 in df[events]:
+            if str(event1) == str(event2):
+                continue
+            # if _locally_happens_directly_before(df, event1, event2, hb, config):
+            if _happens_directly_before(df, event1, event2, hb):
+                # compute the correlation between e2e latency and event1-event2 latency
+                l3 = list(df[event2] - df[event1])
+                if any(map(lambda x: x != 0, l3)):
+                    correlation = pearsonr(l3, e2e)[0]
+                else:
+                    correlation = 0
+
+                hdb += [{'Start': str(event1), 'End': str(event2), 'Correlation': correlation}]
+
+    cfg = _plot_controlflow_graph(df, hdb)
+    _plot_critical_regions(hdb)
+    return LatencyAnalysis(cfg=cfg, hdb=hdb)

xlap/command_line.py

@@ -43,7 +43,8 @@ def main():
                 print(output)
         elif command == "latency":
             df_data = evaluate(data_files["sender"], data_files["receiver"], config=config, kind=0)
-            latency.analyse(df_data, config)
+            a = latency.analyse(df_data, config)
+            print(a.corr.sort_values(ascending=False))
         else:
             df_data = evaluate(data_files["sender"], data_files["receiver"], config=config, kind=0)