Date of Publication : April

Abstract:Cloud computing has become the backbone of modern digital infrastructure, with enterprises increasingly adopting multi-cloud strategies to achieve scalability, fault tolerance, redundancy, cost efficiency, and vendor independence. However, this shift introduces major challenges in orchestration, unified monitoring, threat analysis, and maintaining security visibility across heterogeneous cloud providers, where security data often remains siloed. At the same time, cyber threats continue to evolve in scale and sophistication, requiring systems that are scalable, resilient, and capable of detecting and analysing malicious activity in real time. Honeynets provide controlled environments for observing real-world cyber-attacks and extracting valuable threat intelligence by capturing attacker behaviour, tactics, and tools. Despite their benefits, most existing honeynet deployments remain limited to single-cloud or on-premise environments, resulting in fragmented intelligence, poor scalability, and difficulty processing large volumes of distributed attack data. This paper proposes the design and implementation of a cloud-native, multi-cloud honeynet framework integrated with a scalable data pipeline for centralized data ingestion, enrichment, analysis, and visualization. Low-interaction honeypots deployed across Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) capture malicious activity, which is aggregated through an automated pipeline using cloud-native services such as serverless compute, logging systems, and message queues. Collected data is enriched using external cyber threat intelligence platforms including AbuseIPDB, VirusTotal, Shodan, and AlienVault OTX, and further processed through automated attack classification techniques. Interactive dashboards, time-series analytics, and global attack maps provide real-time visualization of adversary activity, improving situational awareness for defenders. Beyond its cybersecurity application, the proposed framework demonstrates how multi-cloud architectures and data engineering principles can be combined to deliver scalable, resilient, and intelligence-driven monitoring systems. The results contribute both as a proof-of-concept honeynet platform and as a broader model for reliable multi-cloud data analytics.

Reference :