Slow plotting of example script (line10k.html) by Chrome

Hi,

System used: Intel Core i5-3340 CPU @ 3.10GHz with 4.00 GB RAM, 64-bit Win7 and the browser is the up to date Chrome Version 58.0.3029.110 (64-bit).

I loaded Accelerating with WebGL — Bokeh 2.4.2 Documentation from Chrome. The page loaded up instantaneously. As an experiment, I copied the last script of the page that produces “line10k.html” to Jupyter Notebook and time it. Chrome took 22s to plot the 12,705 KB html file produced. Why is the output file so large? See attachment. I don’t think the slowness is due to Chrome flag settings because the user_guide/webgl.html was fast on the same Chrome instance. I have attached the flags here just in case you might need it to see my environment. I have tested other scripts from the user_guide, but same thing, it is slow. Can someone please help? Thank you.

line10k.zip (807 KB)

flags.htm (520 KB)

I found the solution posted by Bryan Van de ven in topic “What could cause a smaller standalone html file to take longer to load?”.

The solution is: add two lines below to the script.

from bokeh.io import reset_output

reset_output()

Thanks to Bryan.

Gss

···

On Tuesday, May 30, 2017 at 9:36:45 PM UTC+8, Gss G wrote:

Hi,

System used: Intel Core i5-3340 CPU @ 3.10GHz with 4.00 GB RAM, 64-bit Win7 and the browser is the up to date Chrome Version 58.0.3029.110 (64-bit).

I loaded http://bokeh.pydata.org/en/latest/docs/user_guide/webgl.html from Chrome. The page loaded up instantaneously. As an experiment, I copied the last script of the page that produces “line10k.html” to Jupyter Notebook and time it. Chrome took 22s to plot the 12,705 KB html file produced. Why is the output file so large? See attachment. I don’t think the slowness is due to Chrome flag settings because the user_guide/webgl.html was fast on the same Chrome instance. I have attached the flags here just in case you might need it to see my environment. I have tested other scripts from the user_guide, but same thing, it is slow. Can someone please help? Thank you.