About the Project – QuantaraX

QuantaraX – “When network fails, Quantara prevails.”

QuantaraX is a fast, resilient, and secure file transfer application designed to tackle unstable and unreliable networks. In situations where timely and accurate data is critical from healthcare to disaster management even a brief connection drop can have severe consequences. QuantaraX ensures no data is lost, corrupted, or left undelivered, providing data integrity, high throughput, and seamless recovery through an adaptive, intelligent transfer engine.

QuantaraX was motivated by two real-world annoyances:

Kaggle Dataset Download: While doing a college project, I was unable to download a 200GB dataset. Browser downloads would repeatedly fail because of spurious connections and throttling bandwidths. One senior team member introduced me to Aria2, an open-source download client that makes use of parallel segmented downloads, automatic retries, and adaptive bandwidth allocation. Aria2's smart handling of failures with concurrency and segmentation raised a question in my mind: why not use this for peer-to-peer transfers?
Direct P2P Transfer: Later, I needed to transfer a large dataset from a friend’s laptop. Cloud solutions were slow and had file size limits; conventional P2P tools failed, forcing us to resort to physical drives. This highlighted the need for a resilient, adaptive P2P system that guarantees delivery without complex configurations.

What it Does

QuantaraX facilitates safe, rapid, and reliable file transfers over unreliable networks:

Adaptive Chunking Engine: Files are broken down into chunks $c_1, c_2, \dots, c_n$ with 256KB–8MB, optimally sized depending on throughput, RTT, and packet loss. Each chunk is signed using SHA256 and HMAC, and encrypted using AES-256.
Parallel QUIC Streams: Transmissions are carried out over 2–8 QUIC streams in such a way that greater than one chunk is transmitted at a time, and bandwidth is utilized to the fullest.
Adaptive Controller: Dynamic observation of parameters and control over chunk size $c_i$ and stream number $n$ by:

$$(c_i, n) = f(\\{throughput, RTT, loss})$$

Priority Scheduling: Critical files receive up to 80% bandwidth so that critical data reaches first.
Retry Mechanisms: Using exponential backoff to retry failed chunks automatically.
User Experience: Directly linking devices through encrypted URLs or QR codes, with a real-time speed, retry, and verification status dashboard.

How We Built It

Core Engine: Go, for high-concurrency performance, QUIC, chunking, encryption, and adaptive control.
Frontend: React + Tailwind, real-time dashboard through Socket.io.
Networking: quic-go for QUIC streams, gorilla/websocket for relay fallback.
Security: SHA256, HMAC, AES-256 encryption.
Pairing: UUID tokens with QR code generation (skip2/go-qrcode).
Storage: Local JSON manifests to store chunk metadata and retries.

It combines Aria2-inspired adaptive chunking with secure P2P downloading, the entire optimized for real-world network instability.

Challenges We Ran Into

Optimal throughput through chunk size and stream number balance without network clogging.
NAT/firewall traversal without third-party VPNs or manual configuration.
Creating a responsive real-time dashboard without any adverse effect on transfer performance whatsoever.
Delivering data security at the expense of none of the speed.

Accomplishments That We're Proud Of

Created an adaptable, strong file transfer system that could work with flaky networks.
Exploited priority scheduling and automatic retries, guaranteeing priority file transmissions.
Introduced a cross-platform dashboard with real-time statistics, easier to believe transfers in real time.
Combined Aria2-style parallelism with P2P transfers, achieving efficiency other approaches lacked.

What We Learned

Advanced networking concepts: QUIC, UDP, congestion control, NAT traversal, and relay protocols.
Adaptive systems and feedback loops for dynamic performance tuning.
Integration of security measures (hashing, HMAC, encryption) without sacrificing speed.
Importance of user-centered design: even complex systems must be intuitive and reliable.

What's Next for QuantaraX

Implement automatic relay fallback in all real-world scenarios.
Explore dynamic stream scaling beyond 8 streams for high-bandwidth environments.
Introduce multi-device simultaneous transfers for collaborative teams.
Optimize for mobile-first deployment in remote healthcare and disaster relief contexts.
Investigate AI-driven adaptive tuning for predictive network adjustments.