Their architecture is shit and done with noob web developer skills. They probably patch it, but last 3months traffic hit them harder and now they are reworking their code. There are massive noob developer mistakes. For example, why there is almost nothing in the client. If you want to make trade during lag, you have to wait for your balance to load from the server. Like you werent there 1min ago and it changed. First. They need to cache all balances and orders in the client. Second, I dont have to wait all charts to load to close order or put one. I already know the numbers I want to enter for that order. Polo wont allow your client to go on and will block you with spinner. They need to remove that blocking mechanism. It only blocks newbie users who don't use apis or client patches like this one https://steemit.com/poloniex/@infu/poloniex-lag-quick-fix
Nomatter how many servers they buy if they made their architecture to have like one nodejs, which is bottleneck in communication, they will have to rewrite everything there to fix the issue. Even buying servers for 5bil wont help.
And they are actually cloud hosted. If you know what you are doing, baremetal is 10x times faster then clouds. Obvously they don't and expected that their performance issues will be solved by dragging the performance slider on a cloud and paying more. Obviously it didn't work. Performance needs to be solved in the application layer which needs to use the hardware optimally. I could write a trading system with more troughtput using one nVME in a .bat file for scripting the trades. Ofc nobody will know about it or use it :)
RE: POLONIEX is basically BROKEN these days. How long until people stop using it?
You are viewing a single comment's thread from:
POLONIEX is basically BROKEN these days. How long until people stop using it?
I think they should hire you!
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit