- What are their security models?
- How are they protecting your data? Is it replicated to multiple destinations?
- How are they storing your data? Is it encrypted, and how are the keys protected?
- Are they using your data for something they shouldn't?
- How would large restores be handled?
SLA checklist:
- Security and Privacy (see 1 & 4 above)
- Redundancy (see 2 above)
- Bandwidth (see 5 above)
- Response time (more of a managed service item, as in "how long to initiate a restore request")
- DR - service provider's ability to recover from their own disaster, and service levels should match your RTO/RPO.
- Non-performance clauses - rebates, waive fees, termination after multiple failures
- Access to service provider data facility
- Certification (SAS70 or PCI)