Soru

A biomedical research facility processes and stores genomic data in a single Google Cloud region ($us-central1$). The organization generates $500\text{ TB}$ of raw genomic sequencing files (FASTQ format) annually alongside structured variant analysis records. The raw sequencing files are accessed heavily during the first 30 days of initial pipeline analysis, queried infrequently between 30 and 365 days, and must be retained for 7 years to satisfy regulatory compliance. The structured variant analysis dataset requires traditional ACID-compliant relational transactions, operates entirely within the single region, and experiences a peak load of 1,500 queries per second. You need to design an infrastructure architecture that minimizes ongoing operational and storage costs while meeting all performance, transactional consistency, and compliance mandates. Which TWO architectural decisions should you recommend?