Misc/speedup_plot at main · AlexanderDBolton/Misc · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt

# before and after bar chart
df = pd.DataFrame({
    "description": ["visits", "search", "search2", "hotel"],
    "before": [130, 230, 245, 12],
    "after": [15, 34, 23, 3]
})

df_melted = df.melt(id_vars="description", var_name="phase", value_name="time")

plt.figure(figsize=(6,4))
sns.barplot(data=df_melted, x="description", y="time", hue="phase", palette=["#d95f02", "#1b9e77"])
plt.yscale("log")  # optional: makes differences clearer
plt.ylabel("Runtime (s)")
plt.title("Query Runtime Speedup")
plt.tight_layout()
plt.show()

# speed-up
df["speedup"] = df["before"] / df["after"]

plt.figure(figsize=(6,4))
sns.barplot(data=df, x="description", y="speedup", color="#1b9e77")
plt.ylabel("Speedup (×)")
plt.title("Query Speedup Factors")
plt.tight_layout()
plt.show()

# dumbbell
import matplotlib.pyplot as plt

plt.figure(figsize=(6,4))
for i, row in df.iterrows():
    plt.plot([row["before"], row["after"]], [i, i], color="gray", linewidth=2)
    plt.scatter(row["before"], i, color="#d95f02", s=100, label="Before" if i==0 else "")
    plt.scatter(row["after"], i, color="#1b9e77", s=100, label="After" if i==0 else "")
plt.yticks(range(len(df)), df["description"])
plt.xlabel("Runtime (s)")
plt.legend()
plt.title("Query Runtime Reduction")
plt.gca().invert_yaxis()
plt.tight_layout()
plt.show()