Update readme with longer example

Author: Rperry2174

examples/debug_python_with_pyroscope.md

@@ -23,10 +23,11 @@ During the period of 100% CPU utilization, you can assume:
 
 The question is: **which part of the code is responsible for the increase in CPU utilization?** That's where flame graphs come in!
 
-## How to use flame graphs to debug performance issues and save money
+## How to use flame graphs to debug performance issues (and save $66,000 on servers)
 Let's say the flame graph below represents the timespan that corresponds with the "incident" in the picture above where CPU usage spiked. During this spike, the server's CPUs were spending:
 - 75% of time in `foo()`
 - 25% of time in `bar()`
+- $100,000 on server costs
 
 ![pyro_python_blog_example_00-01](https://user-images.githubusercontent.com/23323466/105620812-75197b00-5db5-11eb-92af-33e356d9bb42.png)
 
@@ -41,7 +42,7 @@ In this case, `foo()` is taking up 75% of the total time range, so we can improv
 ## Creating a flame graph and Table with Pyroscope
 To recreate this example with actual code, we'll use Pyroscope - an open-source continuous profiler that was built specifically for debugging performance issues. To simulate the server doing work, I've created a `work(duration)` function that simulates doing work for the duration passed in. This way, we can replicate `foo()` taking 75% of time and `bar()` taking 25% of the time by producing this flame graph from the code below:
 
-![image](https://user-images.githubusercontent.com/23323466/105621021-f96cfd80-5db7-11eb-8ceb-055ffd4bbcd1.png)
+<img width="897" alt="foo_75_bar_25_minutes_30" src="https://user-images.githubusercontent.com/23323466/105665338-acf2f200-5e8b-11eb-87b7-d94b7bdda0fc.png">
 
 
 ```python
@@ -61,25 +62,24 @@ def bar():
 ```
 Then, let's say you optimize your code to decrease `foo()` time from 75000 to 8000, but left all other portions of the code the same. The new code and flame graph would look like:
 
-![image](https://user-images.githubusercontent.com/23323466/105621075-a9db0180-5db8-11eb-9716-a9b643b9ff5e.png)
+<img width="935" alt="foo_25_bar_75_minutes_10" src="https://user-images.githubusercontent.com/23323466/105665392-cd22b100-5e8b-11eb-97cc-4dfcceb44cdc.png">
 
 ```python
-# where each iteration simulates CPU time
-def work(n):
-    i = 0
-    while i < n:
-        i += 1
-
 # This would simulate a CPU running for 0.8 seconds
-def a():
+def foo():
     # work(75000)
     work(8000)
 
 # This would simulate a CPU running for 2.5 seconds
-def b():
+def bar():
     work(25000)
 ```
+## Improving `foo()` saved us $66,000
+Thanks to the flame graphs, we we're able to identify immediately that `foo()` was the bottleneck in our code. After optimizing it, we were able to significantly decrease our cpu utilization.
+
+![image](https://user-images.githubusercontent.com/23323466/105666001-1a535280-5e8d-11eb-9407-c63955ba86a1.png)
+
+
+This means your total CPU utilization decreased 66%. If you were paying $100,000 dollars for your servers, you could now manage the same load for just $34,000. 
 
-This means your total CPU utilization decreased 66%. If you were paying $100,000 dollars for your servers, you could now manage the same load for just $66,000. 
 
-![image](https://user-images.githubusercontent.com/23323466/105621350-659d3080-5dbb-11eb-8a25-bf358458e5ac.png)