rafMetrics

Now published: A new metric for evaluating the performance and complexity of computer programs: A new approach to the traditional ways of measuring the complexity of algorithms and estimating running times at the 2021 23rd International Conference on Control Systems and Computer Science (CSCS).

@INPROCEEDINGS{9481065, author={Folea, Rares and Slusanschi, Emil-Ioan}, booktitle={2021 23rd International Conference on Control Systems and Computer Science (CSCS)}, title={A new metric for evaluating the performance and complexity of computer programs: A new approach to the traditional ways of measuring the complexity of algorithms and estimating running times}, year={2021}, volume={}, number={}, pages={157-164}, abstract={This paper presents a refined complexity calculus model: r-Complexity, a new asymptotic notation that offers better complexity feedback for similar programs than the traditional Bachmann-Landau notation, providing subtle insights even for algorithms that are part of the same conventional complexity class. The architecture-dependent metric represents an enhancement that provides better sensitivity with respect to discrete analysis.}, keywords={}, doi={10.1109/CSCS52396.2021.00033}, ISSN={2379-0482}, month={May},}

Metrics for evaluating the performance and complexity of computer programs.

• Written Material: here
• Presentation: here
• Short Material: here

Part of Bachelor Thesis @ CS - Full Material: here - Presentation (ro): here - Short demo: link

Components:

rafComputing

The ML-based tool implementing the process of automatically tailoring a suitable rComplexity Class for an algorithm.

NOTE: Improved version: https://github.com/raresraf/rafPipeline/tree/master/rComplexity

WebMonitoring

A tool for monitoring multiple network resources and websites. Gather data by periodically monitoring specific resources and websites and stores results in database.

ResourceManager

Monitors all resources by periodically (timer set default at 1 hour interval) sending requests to existing resources. Store simple metrics like total time or total requests answer as entries in DB.

WebsiteManager

Monitors all websites by periodically (timer set default at 1 hour interval) generating a HAR (HTTP-Archieve data performance file) for loading metrics corresponding to a website, with Chrome using Browsermob-Proxy. Also parse and store valuable insights resulted from the HAR file into DB. The service uses speedprofile engine.

WebMonitoring API

Provide an API for interrogating useful metrics from DB.

Login

Backend implementation to provide a simple authentication, registration and management for users inside rafMetrics platform.

DockerConfig

Keeps track of all Docker Compose/Docker Swarm settings and configurations.

KubernetesConfig

Keeps track of all k8s settings

MySQL

Database used to store persistent data required by Login and WebMonitoring. All relations are kept in Boyce-Codd Normal Form.

deploy_repo.sh

Simple script to ensure dockerize and deployment in Kubernetes for all backend components

metricsUI

Frontend implementation of rafMetrics platform based on Flatlogic Template: React Material Admin — Material-UI Dashboard