antani

Ant - agent/network intelligence

antani_logo
antani_logo

ants optimizing paths on a network

workflow

The optimization engine has user controlled and automated modules

ops_app
ops_app

ops application

We retrieve information directly from the fleet console microservices

man_console
man_console

management console

We collect the current status with a single api call

resq = requests.get(url=cred['node_url']+"/users/actions/",headers=headers)
resq = requests.get(url=cred['node_url']+"/geodata/zones/",headers=headers)
resq = requests.get(url=cred['node_url']+"/geodata/zones/"+id_zone+"/areas",headers=headers)
resq = requests.get(url=cred['node_url']+"/geodata/areas/deployment/zone",headers=headers,params={"zoneIdentifier":g['zoneIdentifier']})

The fleet engine sits in the middle and steers all operations, tracking and monitoring log the current status

obj_dev
obj_dev

development of objectives

In the fleet engine 4 objects will be defined with an hierarchical structure

metric_ops
metric_ops

metric operations

The purpose of the optimization engine is to compute profitability for each drive and iterate all over the possible solutions

task preparation
task preparation

field operation assignement

For each drive is calculated a cost and a rating

For Each task is calculated a revenue and a risk

design

To enable parallel work we need to move from a graph design

engine_design_old engine design, current scheme

To a linear design where the engine just orchestrate the calls and there is always a cached table to substitute a broken service

engine_design engine design, suggested scheme

We draw a cut between field operation and task optimization

infra_design infra design