Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

# Enhanced Automatic Relay Selection

IP.com Disclosure Number: IPCOM000248544D
Publication Date: 2016-Dec-15
Document File: 1 page(s) / 19K

## Publishing Venue

The IP.com Prior Art Database

## Abstract

This article will present a formula for selecting best machine / Relay: Enhanced Automatic Relay Selection - alghoritm that uses an equation, that in result will use Delay and Load metric (to determine best Relay to connect to) and ... Luck.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

1

Enhanced Automatic Relay Selection

Enhanced Automatic Relay Selection

Definitions: Endpoint - machine that must decide which Relay to select, from all possible Relays, to connect to (no matter what reason or technology used). Relay - machine (doesnt matter technology) that servers as middleman between Endpoint and "something" else. Relay can handle only X Endpoints (lets say 1000).

Problem: Endpoint must decide, which Relay to use, for optimal connection. Also we must ensure that Relays are not overloaded, that Endpoint are spread even if they are forced to connect to not optimal, but still good Relays. Regular algorithm using Load and Delay will load best Relays to maximum, while worse Relays are not used at all. Thats why a factur is introduced referred to as "Luck".

Equation: metric = Delay x Load x Luck Delay - represent delay, biggest Delay = worse metric, Delay can be measured, by Endpoint, as difference between sending and receiving response from Relay. Load - of Relay, more load = worse metric, Load = (Count of Children) / (Max Child Count, default 1000) Luck - of Relay, every time Endpoint begins ARS algorithm, Endpoint "roll a dice", for every Relay separately, and receive random number between X and Y that represent %. For example, if X = 1 and Y = 100, its a roll between 1% and 100%.