I85PREP
⌘ K
Dell Data Protection Foundations study materials
Dell Data Protection Foundations study resources

Dell Data Protection Foundations: Data Protection Design, Sizing, and Best Practices

Dell Data Protection Foundations · Work through your exam, one topic at a time.

All guide sections

Data Protection Design, Sizing, and Best Practices

Domain 9 tests design and sizing: turning requirements into a protection architecture, calculating capacity and throughput, and avoiding oversizing or undersizing. Expect scenario questions with numbers.

Core rules

  • Start with RPO, RTO, retention, and growth. Then map workloads to protection methods: snapshots, backup, replication, archive, or vault.
  • Capacity sizing uses full size, change rate, backup frequency, retention, dedupe ratio, compression, metadata, and overhead. Dedupe ratio is workload-specific; use a proof of concept or vendor sizing tool rather than a generic assumption.
  • Throughput sizing uses change data divided by the backup window. Add headroom for streams, metadata, verification, and concurrent jobs. A single stream may not saturate the link.
  • Replication bandwidth is based on change rate and RPO, not full capacity. Include overhead and latency.
  • Resiliency requires redundant components, multiple sites, and tested failover. Scalability requires headroom for growth and additional workloads.
  • Security by design includes immutability, encryption, least privilege, and an isolated copy.
  • Documentation includes the architecture, bill of materials, runbooks, test results, and assumptions. A design without assumptions cannot be validated.

Worked example

A 100 TB database has a 10% daily change rate, a 6-hour backup window, and 30-day retention. Daily change is 10 TB. Required throughput is 10 TB / 21,600 s = about 463 MB/s before dedupe. If measured dedupe is 4:1, the daily unique data is about 2.5 TB, but the backup stream still carries 10 TB unless source dedupe is used. With DD Boost, source dedupe reduces LAN/WAN traffic; target dedupe reduces capacity. A 30-day retention estimate is not 30 x 100 TB; it is the full plus 30 days of unique changes, plus metadata and overhead. A PoC validates the dedupe ratio and stream count.

Traps that cost marks

  • Sizing capacity from full backups only.
  • Using a generic 10:1 dedupe ratio.
  • Ignoring metadata, verification, and growth.
  • Sizing replication for full capacity.
  • Assuming one stream saturates the network.
  • Designing without a tested restore.