Performance and Security Challenges in Cloud-Based Data Storage Systems: A Secondary Empirical Analysis of Object-Storage Latency, Throughput, Availability, and Cyber Risk

Cloud-based data storage offers elastic capacity, geographic redundancy, and pay-per-use access, but organizations must balance performance variability with security exposure. This study synthesizes secondary evidence from three performance studies, official reliability documentation from Amazon Web Services, Google Cloud, and Microsoft Azure, and authoritative threat reports from NIST, the Cloud Security Alliance, Verizon, and ENISA. The analysis examines request latency, throughput, concurrency, durability, availability, misconfiguration, identity and access management, vulnerability exploitation, ransomware, and data exfiltration. In a one-million-request experiment using 1 KiB objects, S3 Standard recorded median read and write latencies of 27 ms and 40 ms, compared with 5 ms and 8 ms for S3 Express, representing reductions of 81.5% and 80.0%. An eight-week object-storage study reported per-object bandwidth of approximately 25–95 MiB/s and aggregate throughput of 75–90 Gbit/s under high concurrency. Security evidence identifies misconfiguration, weak access control, vulnerability exploitation, and ransomware as major risks. The study proposes a performance-security co-optimization framework combining workload-aware storage selection, bounded asynchronous I/O, least-privilege access, encryption, immutable recovery copies, continuous configuration assessment, and centralized observability. The findings show that durability, availability, performance, and security are distinct but interdependent properties that should be engineered and governed jointly.