This Cloud Hosting Service Level Agreement and Technical Specification Matrix is issued by the contracting entity designated as the Provider within the Master Subscription Agreement (MSA) and forms an immutable, integrated component thereof. All technical benchmarks, infrastructure commitments, and availability targets detailed below are guaranteed exclusively by the Provider.
Uptime - 99.9%
Regional Sovereignty: Customer databases are hosted exclusively within the chosen relyd infrastructure deployment zone: Europe, Americas, Asia-Pacific, Middle-East & Northern Asia, or Oceania.
Uptime Commitment: The Provider, via the enterprise-grade hosting architecture of relyd Inc., guarantees a minimum monthly uptime target of 99.9% (three nines), calculated exclusive of pre-scheduled maintenance windows. In the event of a breach of this baseline, any applicable service credits shall be issued solely by relyd Inc. as a deduction against future subscription invoices.
Downtime Threshold: A 99.9% monthly availability target permits a maximum of 43 minutes and 49 seconds of unplanned downtime within any single calendar month.
Operational Performance: In practice, our robust infrastructure backbone consistently exceeds these baselines, as our underlying tier-one hosting providers historically outperform their own contractual SLA benchmarks.
High Availability
Infrastructure Resilience: Data centres are Tier-III certified or equivalent, utilizing a minimum of N+1 redundancy across all power distribution and network routing architectures to eliminate single points of failure.
Real-Time Replication: Each production database is replicated in near real-time onto redundant storage systems located entirely within the same chosen geographical deployment zone.
Failover & Data Integrity: This synchronous architecture enables rapid, seamless failover mechanisms in the event of any hardware disruption, ensuring maximum operational continuity with minimal to no data loss.
Backups & Disaster Recovery
Retention Architecture: Backup protocols maintain fourteen (14) full snapshots distributed over a minimum three-month cycle, precisely structured as one daily backup for seven (7) days, one weekly backup for four (4) weeks, and one monthly backup for three (3) or more months.
Geographical Replication: Backups are systematically replicated across a minimum of three (3) distinct physical machines situated in independent data centres across Europe and Canada, supplemented by a fourth, immutable "frozen" copy in isolated cold storage; please note that replication paths are programmatically determined and cannot be geographically restricted by the client.
On-Demand Data Export: Clients retain the sovereign right to generate and download manual backups of their live production data at any given time via the control panel.
Single Server Disruption Targets: In the event of a permanent localized hardware failure affecting an isolated server, the Disaster Recovery Plan dictates a Recovery Point Objective (RPO) of a maximum of twenty-four (24) hours - historically trending near-zero for database workloads - and a Recovery Time Objective (RTO) ensuring service restoration within six (6) hours.
Data Centre Catastrophe Targets: In the event of a total and permanent data centre outage, the Disaster Recovery Plan dictates an RPO of twenty-four (24) hours of maximum potential data variance, and an RTO of twenty-four (24) hours to completely provision and restore the environment within an alternative, unaffected data centre.
Security
Sovereign Data Protection: The absolute safety of client data remains the paramount priority of relyd Inc.; all infrastructure systems, tenant isolation parameters, and operational procedures managed by the Provider are systematically designed to guarantee comprehensive data security under applicable global frameworks.
Cryptographic Architecture: Customer data is fully encrypted at rest utilizing industry-standard cryptographic methods, as well as in transit via 256-bit Transport Layer Security (TLS/HTTPS) backed by robust 2048-bit modulus SSL certificates and regularly audited configuration sets.
Infrastructure Resilience: Database environments leverage tier-one data centres featuring full hardware guarantees alongside comprehensive N+1 redundancy across all storage, network routing, and electrical power supplies.
Credential Security: Authentication credentials are protected utilizing industry-standard PBKDF2 with SHA-512 hashing protocols, structurally strengthened via unique cryptographic salting and stretched for thousands of algorithmic rounds to eliminate brute-force vulnerability.
System Hardening: Production environments operate on minimal, hardened, and standardized system images derived from modern Linux distributions, featuring automated unattended security patching, active firewalls, and advanced intrusion countermeasures.
Tenant Isolation: Complete structural isolation is enforced across the platform; customer data is hosted within dedicated databases, strictly prohibiting data sharing or cross-database access between discrete client environments.