BitShares Forum
Main => General Discussion => Topic started by: litepresence on June 06, 2018, 01:52:45 pm
-
metaNODE = Bitshares_Trustless_Client()
There were two ways to get data from Bitshares blockchain:
- a private node that uses lots of RAM, prefers its own machine, and is technical to tend
- a public node that is difficult to stay connected to and may provide rogue data
I've created a 3rd path; connect to several random nodes in the public network... ask them each for latest market data; then move on to other nodes in the network continually. Finally, perform statistical analysis on the combined feeds and maintain a streaming curated output file; the metaNODE.
python script and whitepaper, including usage:
https://github.com/litepresence/extinction-event/tree/master/metaNODE
LICENSE:
'litepresence 2018'
def WTFPL_v0_March_1765():
if any([stamps, licenses, taxation, regulation, fiat, etat]):
try:
print('no thank you')
except:
return [tar, feathers]
Mission:
1/10 the RAM usage of a personal node
99.9999 six sigma uptime
99.9999 six sigma accurate data feed
less than 5000ms latency
metaNODE is a streaming text file that contains statistically curated dex data from all public nodes for a single market on a single account. metaNODE currently curates the following feeds:
- last
- market history
- open orders
- orderbook
- market-account balance
Run time has been demonstrated in excess of 2 weeks.
metaNODE.txt is updated about once per second with live data.
metaNODE is entirely independent of pybitshares.
Public database calls are made with websocket-client.
This project has received funding through the DEXbot worker. Additional funding has been earmarked for incorporation of metaNODE into future release of the DEXbot platform.
metaNODE10 has already been incorporated into the Extinction Event dex algo trading framework EV.py; learn more at
www.litepresence.com
-
Nice work! If you can create a nodes' data health monitoring web page, then everyone can check which node is good and which node is bad.
-
I think knowing which nodes are bad + monitoring will be very important to traders +1 Bangzi
-
I've created a 3rd path; connect to several random nodes in the public network... ask them each for latest market data; then move on to other nodes in the network continually. Finally, perform statistical analysis on the combined feeds and maintain a streaming curated output file; the metaNODE.