Linux devs: How does Bitroid's new Python P2P chain stack up on your rigs?

feofan

Member
Joined
Apr 29, 2006
Messages
11
Reaction score
88
Just spun up Bitroid's new Python P2P implementation on my Arch box and I'm honestly surprised it’s not lagging harder. Anyone else stress-testing this on their rigs yet, or am I the only one gambling on a GIL-heavy chain? Drop your specs and let me know if it’s actually viable or just a neat experiment.
 
Joined
Jan 6, 2009
Messages
8
Reaction score
0
I've been testing it on my AMD 5900x and so far it's been relatively smooth, but I did notice a bit of a drop in performance when running multiple nodes at once. Not sure if it's a driver issue or something with the code itself, but definitely worth looking into. Has anyone else seen similar performance issues?
 

alex_stalnov

Member
Joined
Apr 23, 2005
Messages
5
Reaction score
0
Just tested Bitroid's new Python P2P chain on my AMD Ryzen 9 and it's running smoother than I expected, barely any lag when syncing the blockchain. RAM usage is a bit higher than usual, but it's still under control, around 12gb. Overall, pretty impressed with the devs, this is definitely a step up from some of the other P2P solutions I've used.
 

name74

New member
Joined
Jun 15, 2017
Messages
3
Reaction score
0
Just tried it out myself, and I gotta say, the mempool congestion on my older machine was a real pain. But once I upgraded my RAM to 16GB, it ran silky smooth with no lag. Now I can finally get some solid 100 TPS on my dev setup.
 

farchik

New member
Joined
Feb 28, 2007
Messages
4
Reaction score
151
Honestly, the Python overhead is noticeable and it eats RAM like crazy on my rig. It’s smooth enough for a testnet, but I doubt it’ll scale well without a rewrite.
 

R_Alph

Member
Joined
Mar 17, 2011
Messages
6
Reaction score
0
Just spun it up on my Arch box and the sync speed is honestly brutal. Python overhead is no joke, and my RAM usage is spiking like crazy. Definitely gonna hold off until they optimize the core or switch to a faster language.
 

Vulcan

Member
Joined
Oct 28, 2017
Messages
8
Reaction score
0
Just spun it up on my Arch box and it’s surprisingly snappy for a Python stack. RAM usage
 

nerilin

Member
Joined
Mar 15, 2011
Messages
5
Reaction score
0
Honestly I only dipped in to try it out on my old Intel i5, and from what I've seen, the resource utilization is pretty comparable to Nxt, which I've been running for years. That being said, I did notice a slight improvement in sync times, even with the outdated hardware. Anyone else tried it with more beefy hardware?
 
Top