Aggregated STEEM flows intensity statistics from last 2 weeks:

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/steem_flow161106.svg.png]
Legend:
- spm - STEEM per minute
- $pm - SBD per minute

This is the second post that is continuing issue from my previous post. Now steem_flow2.py script is capable to store calculated data in Redis DB as I promised in previous post. Also it is possible to display stats over web-server and generate statistics graphics per day.

STEEM blockchain transaction statistics tool:
* calculate average STEEM & SBD flows intensity and ratio;
* store data in database (Redis DB);
* display stored data over web-server (Twisted framework);
* plot statistics graphics (Pandas, Matplotlib).

Daily STEEM/SBD flows statistics:

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/steem_ex.png]

As we see STEEM flow in direction to exchanges prevails and press STEEM ex-rate down.

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/sbd_ex.png]

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/flow_ratio.png]

SBD flow to/from exchanges is almost the same in average.

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/convert_sbd.png]

SBD convert flow is more intensive than exchange flow and sometimes we could observe extremely high peak. SBD to null flow (post promotion) is not significant.

[IMAGE: https://github.com/fooblic/steem-flow/raw/master/steem_power.png]

As usual power down is several times more intensive than power up with withdrawals occasional peaks.

All source codes could be found on github: https://github.com/fooblic/steem-flow

Further I plan to prepare aggregated statistics on weekly and monthly base as well.