PLATCOM
A Platform-Ready Data Compression Solution for Optimized Big Data Storage
Project Summary:
PLATCOM is an innovative, platform-ready data compression solution designed to address
the growing demands of big data management across industries. By leveraging advanced
multi-scheme compression algorithms, PLATCOM optimizes data storage and retrieval,
reducing infrastructure costs and enhancing data processing efficiency. The tool is
particularly suited for sectors that generate and handle vast data volumes, including
telecommunications, finance, healthcare, and public sector organizations involved in
smart city initiatives.
The project will proceed in distinct phases, focusing initially on evaluating and
incorporating new compression algorithms into PLATCOM and then on rigorously testing
PLATCOM's performance with real and synthetic datasets. An agile methodology will be
adopted to ensure flexibility and continuous improvement of the machine learning
algorithms for data clustering and suitable compression algorithm selection, with
insights from testing guiding iterative refinements.
PLATCOM aims to deliver a market-ready Minimum Viable Product (MVP) by the project's
end, certified and compliant with industry standards. Additionally, it will be equipped
with essential features such as advanced data retention options and customizable
compression profiles, enhancing its adaptability across various use cases.
Key objectives include securing intellectual property rights, generating strategic
partnerships, and conducting pilot deployments with early adopters to gather valuable
feedback and further validate PLATCOM's real-world effectiveness. Ultimately, PLATCOM is
positioned to become a cornerstone solution for data-heavy industries, reducing
operational costs, enhancing performance, and supporting data-driven decision-making
across a range of applications.
Details:
| Programme | Creation and Initial Development of Startups with International Orientation |
|---|---|
| Proposal Number | PRE-SEED/0824/0230 |
| Proposal Acronym | PLATCOM |
| Funding |
The project is implemented under the programme of social cohesion “THALIA 2021- 2027” co-funded by the European Union, through Research and Innovation Foundation. |
Paper:Rethinking Big Data Scale: Balancing Accuracy, Collaboration, and Storage
The scale of modern big data is increasingly shaped by collaboration as a result of the constant exchange and sharing of information between organizations, users, and systems. As data collaboration grows, the cost of moving, storing and synchronizing shared datasets has become a major challenge. Existing systems focus on capacity and compression within isolated environments, overlooking the broader need to reduce redundant data movement while keeping results accurate and consistent.In this vision paper, we argue that balancing accuracy, collaboration, and storage should be at the core of rethinking the big data scale. We propose an approach where systems learn when and what to share, reducing unnecessary transfers through collaborative compression, shared summaries, and intelligent reconstruction. Rather than sending entire datasets, collaborating sites can exchange compact representations that preserve essential information, while AI methods fill in or refine details when needed. By aligning accuracy with collaboration efficiency, future data systems can scale not by storing or moving more data, but by sharing smarter. We outline key design ideas for such collaboration-aware systems and discuss open challenges in building intelligent, efficient, and sustainable data infrastructures.
PoC: COMPASS is a multiple compression tool utilizing attribute-based signatures. COMPASS exploits K-means clustering to select the best compression scheme for different data subsets in a database. The experimental results show that COMPASS significantly reduces disk space usage compared to monolithic methods. COMPASS
Basic Research: Our preliminary experimental evaluation of our first prototype using three compression algorithms and two real datasets, we observed reduction in storage space, up to ~4% against the competitors. SIBACO