Stretch HA Storage Solution
Multi-Site Active-Active Replication with Geographic Resilience
Advanced stretch clustering delivering synchronous replication across geographic sites. Provides complete disaster recovery, site-level fault tolerance, and continuous operation with zero data loss protection.
When to Choose Stretch HA
Essential for organizations requiring disaster recovery, regulatory compliance, or multi-site operations. Ideal when business continuity demands protection against site-level failures including natural disasters, power grid failures, or major infrastructure outages.
Site-Level Protection
Survives complete site failure including power outages, natural disasters, or infrastructure failure with automatic recovery.
Zero Data Loss
Synchronous replication ensures both sites maintain identical data – no transactions are lost during site failures.
Continuous Operation
Services continue seamlessly at surviving site during failures, with automatic service restoration when failed site recovers.
Geographic Flexibility
Deploy across data centers, availability zones, or metro areas with support for up to 100km separation.
Stretch Clustering Architecture
Geographic distribution with synchronous replication across sites. Each location maintains complete infrastructure independence while RSF-1 orchestrates real-time data consistency and automatic failover.
Geographic Distribution
Two controllers deployed across separate sites with dedicated storage at each location, providing disaster recovery and site-level fault tolerance.
Site-Independent Storage
Each site maintains complete N+1 JBOD infrastructure, enabling full operation even during complete site failure or extended network partitions.
Synchronous Replication
Real-time data synchronization across sites ensures both locations maintain identical datasets with transaction-level consistency.
Performance Tiers
Choose the right performance level for your specific workload requirements
Entry Series Stretch
Processor:
2× AMD X-Core CPUs
Memory:
2× 256 GB ECC
Included Features:
Dual OCP Cards per Controller
Site-Independent Storage
Synchronous Replication
100km Maximum Distance
SAS JBOD Configurations
Ideal for:
Distributed file services, geographic backup, disaster recovery for mid-sized deployments
Base Series Stretch
Processor:
2× AMD Y-Core CPUs
Memory:
2× 512 GB ECC
Included Features:
Dual OCP Cards per Controller
Site-Independent Storage
Synchronous Replication
100km Maximum Distance
SAS and NVMe JBOD Support
Ideal for:
Virtualization platforms, departmental storage, mixed workloads
Performance Series
Processor:
2× AMD Z-Core CPUs
Memory:
2× 1024 GB ECC
Included Features:
Dual OCP Cards per Controller
Site-Independent Storage
Synchronous Replication
100km Maximum Distance
NVMe Performance Optimization
Ideal for:
Financial trading systems, real-time analytics, mission-critical multi-site applications
Technical Specifications
Component
Specification
Configuration
2 Controllers and N+1 JBODs (one per site)
Geographic Separation
Controllers deployed at separate sites with independent power and network
Memory per Controller
256 GB / 512 GB / 1024 GB ECC DDR4
Client Network per Site
4×10 GbE or 4×25 GbE (dual OCP cards)
Inter-Site Replication
2×40 GbE or 2×100 GbE dedicated WAN links
Storage per Site
Up to 4 JBOD enclosures (60-bay SAS or 24-bay NVMe)
Maximum Distance
Up to 100km with proper network infrastructure
WAN Requirements
Dedicated fiber with <10ms latency for synchronous replication
Site Failover Time
Sub-30 second automatic site failover
Data Consistency
Synchronous replication ensures zero data loss during site failure
Network Switching
DA or Arista 100 GbE switches for high-speed fabrics
Disaster Recovery
Complete site recovery with automatic service restoration
Disaster Recovery & Site Failover
Site Failure Response
Site failure detected through heartbeat loss and network monitoring
Surviving site assumes all services and activates floating IPs
DNS updates redirect client traffic to operational site
Full service restoration with current data and zero loss
Site Recovery Process
Failed site infrastructure restored and controllers restarted
RSF-1 initiates automatic data resynchronization
Integrity checks verify data consistency across sites
Stretch clustering automatically reestablished
Network Requirements & Considerations
WAN Connectivity
• Dedicated fiber links with <10ms latency
• Minimum 40 GbE bandwidth for replication
• Redundant network paths recommended
• Quality of Service (QoS) for replication traffic
Distance Limitations
• Maximum 100km for synchronous replication
• Metro-area and campus deployments ideal
• Consider asynchronous for longer distances
• Network latency impacts write performance
Ready to Transform Your Storage Infrastructure?
Let’s discuss your requirements and design the perfect solution.