CrowdCloud is a decentralized web platform designed to democratize access to computing and storage resources.
It enables users to borrow or contribute compute power in a distributed, serverless environment — supporting general-purpose workload execution and persistent data storage with built-in incentives.
This repository focuses on the CrowdCloud Web Application, primarily developed for Borrowers (users who request computational jobs).
As global digital workloads grow—fueled by artificial intelligence, big data, and IoT devices—the demand for scalable and affordable compute resources continues to rise.
However, current cloud models remain centralized, leading to high costs, vendor lock-in, and reduced transparency.
CrowdCloud addresses this by introducing a decentralized and serverless platform where users can share idle resources, execute workloads securely, and earn incentives for contributing compute power.
This initiative aims to democratize access to infrastructure, allowing even underfunded or marginalized communities to participate in cloud-scale computation.
To develop a decentralized platform that enables users to contribute and consume serverless compute and storage resources, featuring intelligent workload scheduling, secure isolation, and transparent incentive mechanisms.
- Design and implement a distributed system that allows peer-contributed nodes to execute sandboxed compute jobs.
- Create a distributed blob and object storage layer with redundancy and failover capabilities.
- Develop an AI-based scheduler that intelligently assigns workloads based on real-time resource availability.
- Implement a transparent incentive model using credits or tokens.
- Evaluate system scalability, security, and efficiency through benchmarks and stress tests.
CrowdCloud focuses on the creation of a fully decentralized platform that:
- Executes sandboxed, serverless workloads using lightweight virtualization (e.g., Firecracker/Qemu).
- Provides distributed blob and object storage with replication and fault tolerance.
- Features an AI-powered workload scheduler that optimizes performance and latency.
- Includes a borrower interface (this web app) for managing jobs, analytics, and expenses.
- Provides a lender interface (future phase) for contributors sharing resources.
- GPU and TPU acceleration are not supported in this phase (CPU-only workloads).
- No integration with public blockchains (uses local credit simulation).
- Sandbox limited to Qemu-based isolation.
- Fixed-rate incentive system for prototype evaluation.
- Currently supports desktop and server nodes only — no mobile or edge optimization yet.
The CrowdCloud Web Application serves as the Borrower’s Portal — where users can register, log in, submit compute jobs, and monitor their performance and expenses.
It provides data-driven insights through analytics and visualizations.
- Sign Up – Two borrower types:
- Personal User (individual contributor)
- Company User (individual under an organization)
- Login System – Secure authentication connected to the backend API.
- Overview of resource usage, job activity, and system performance.
- Displays metrics such as:
- Job summary
- Failure rate
- Traffic metrics
- Borrower expenses
- Provides in-depth visualization of borrower data using charts and statistics:
- Request job breakdown
- Active resource utilization
- Monthly cost tracking
- Job status distribution
- Visualized with pie charts, line charts, bar charts, and key metrics.
- Manage and monitor job requests in real-time.
- Organized into three main sections:
- Overview – Summarized job statistics.
- Create Job – Interface for submitting compute jobs with custom resource configurations.
- Job List – Displays active and completed jobs, including CPU and RAM allocations.
CrowdCloud Web Application is currently in active development.
Core borrower-side features (authentication, dashboard, job management, and analytics) are fully functional and connected to the backend.
Pending Implementation:
- Dynamic billing system for borrowed compute and storage resources.
- Billing depends on execution time, workload complexity, and network traffic — currently under research for fair pricing.
Proprietary License — All Rights Reserved
© 2025 CrowdCloud. All rights reserved.
This project and its source code are the intellectual property of the chmod u+rwx.
Unauthorized copying, modification, distribution, or use of this software, in whole or in part, is strictly prohibited without prior written consent.
For collaboration, research partnership, or licensing inquiries, please contact the development team directly.
CrowdCloud envisions a future where everyone contributes to — and benefits from — the global compute cloud.