RED-01
Electrical & Cooling 2N Concurrency
Dual utility transformers, N+1 synchronized generators, 2N A+B UPS buses, Static Transfer Switches (< 4ms), and N+1 precision cooling loops.
Request Specification
DC-05 // REDUNDANCY & HIGH AVAILABILITY
99.999% High-Availability Frameworks, Automated Failover & Disaster Recovery Engineering
Eliminate single points of failure across facilities, power, networks, storage, and databases. Dulo Tech designs and tests end-to-end active-active and active-passive redundancy architectures delivering near-zero RPO and sub-minute RTO.
99.999% Five-Nines Availability
Zero-RPO Synchronous Replication
Multi-ISP BGP & Dark-Fiber Diversity
Automated DR Runbook Orchestration
HIGH AVAILABILITY & DR // 99.999% CONTINUITY
99.999%
Target Five-Nines Uptime (< 5.2m/yr)
0 Sec
Synchronous Metro RPO Data Loss
< 60 Sec
Automated Cluster & DNS Failover
100%
Quarterly Live Failover Drill Pass
ELIMINATING SINGLE POINTS OF FAILURE
True High Availability Means Any Component—or Entire Site—Can Fail Without Stopping Business
Many organizations believe they have redundancy until a real outage strikes—revealing that both fiber ISPs enter through the same street manhole, both UPS units share a single bypass switch, or their backup database hasn't been tested in 18 months.
Dulo Tech performs ruthless Single-Point-of-Failure (SPOF) forensic audits across your physical facilities, electrical buses, network routing, storage fabrics, and application tiers—engineering automated failover systems that are proven through live quarterly chaos drills.
End-to-End Single-Point-of-Failure (SPOF) Elimination Across All 7 OSI & Facility Layers
Synchronous Metro Storage/DB Clustering (Zero RPO) & Asynchronous Cloud DR
Automated One-Click DR Orchestration & Quarterly Live Failover Drills
ZERO-SPOF ARCHITECTURE // ACTIVE-ACTIVE HA
SLA: 99.999%
SIX LAYERS OF REDUNDANCY ENGINEERING
Full-Stack Fault Tolerance From Utility Grid to Application Layer
True 99.999% availability requires synchronized redundancy across all six architectural layers.
RED-01
RED-02
Carrier-Neutral Dark Fiber & BGP Mesh
Dual physically diverse building fiber entries (North/South entry), multi-homing BGP with independent upstream ISPs, and SD-WAN sub-second failover.
Request Specification
RED-03
Core Switching & Firewall HA Clustering
Spine-leaf VXLAN/EVPN fabrics, MC-LAG/VPC dual-homed server NICs, and Active-Active / Active-Passive state-synced firewall pairs.
Request Specification
RED-04
Hypervisor & Storage Metro-Clustering
VMware vSphere HA/DRS, stretched SAN/vSAN metro-clusters across two sites, and dual-controller NVMe arrays with automated VM restart.
Request Specification
RED-05
Database Always-On & Zero-RPO Sync
Oracle Data Guard / RAC, SQL Server Always-On Availability Groups, and PostgreSQL Patroni clusters with automatic read/write leader election.
Request Specification
RED-06
Immutable Ransomware Vault & DR Orchestration
3-2-1-1-0 backup architecture with air-gapped WORM storage and automated Veeam/Zerto one-click Disaster Recovery site runbooks.
Request Specification
HIGH-AVAILABILITY ASSURANCE PROTOCOL
Six Practices That Guarantee Your Failover Actually Works When Needed
Unverified redundancy is an illusion. We prove failover readiness through six continuous practices.
HAP-01
Forensic SPOF Dependency Mapping
Tracing every power cable, fiber patch, DNS record, and authentication dependency to expose hidden common-mode failure points.
SPOF ELIMINATED
HAP-02
Automated Daily Backup SureBackup Boot
Automatically booting your nightly server backups inside an isolated sandbox every morning to verify 100% OS and database recoverability.
DAILY BOOT TESTED
HAP-03
Sub-Minute Global Server Load Balancing
Deploying F5 / Cloudflare / Anycast GSLB health checks that automatically steer users to your DR site if the primary site stops responding.
AUTO DNS GSLB
HAP-04
Quarterly Live Plug-Pull Failover Drills
Conducting scheduled live failover exercises where production traffic is switched to your secondary DR site to satisfy CBK and ISO 22301 auditors.
LIVE DR DRILLS
HAP-05
3-2-1-1-0 Immutable Air-Gap Standard
Maintaining 3 copies of data on 2 media types, 1 offsite, 1 immutable/air-gapped (WORM), with 0 unverified recovery errors.
RANSOMWARE PROOF
HAP-06
ISO 22301 Business Continuity Governance
Aligning technical RTO/RPO metrics with executive crisis communication trees and regulatory reporting mandates.
ISO 22301 ALIGNED
AVAILABILITY & DR TIER MATRIX
Comparison of Resilience Architectures, RPO/RTO & Uptime
Engineering trade-offs and performance benchmarks across Dulo Tech resilience tiers.
| Tier Code | Resilience Architecture | Data Loss Target (RPO) | Recovery Time Target (RTO) | Annual Downtime Allowance |
|---|---|---|---|---|
| HA-T1 | Immutable Offsite Backup (3-2-1-1) | 4 – 24 Hours | 2 – 6 Hours | 99.5% (< 43.8 Hrs / Year) |
| HA-T2 | Warm Standby Cloud / DR Site | 5 – 15 Minutes (Async) | 15 – 30 Minutes | 99.9% (< 8.76 Hrs / Year) |
| HA-T3 | Hot Standby + Auto DB Failover | < 30 Seconds | 1 – 5 Minutes | 99.99% (< 52.6 Min / Year) |
| HA-T4 | Active-Active Metro Cluster | Zero Data Loss (0 Sec Sync) | < 10 Seconds (Seamless) | 99.999% (< 5.26 Min / Year) |
| HA-T5 | Multi-Region Geo-Distributed Active | Zero Data Loss (Consensus) | 0 Seconds (Continuous) | 99.999%+ Mission Critical |
TECHNICAL KNOWLEDGE BASE
Frequently Asked Engineering & Procurement Questions
Direct answers from Dulo Tech Limited's Principal Systems Architects, CSOC Directors, and Institutional Procurement Officers at Mirage Towers, Nairobi.
RPO measures how much data you can afford to lose (in time) before the crash—for example, an RPO of 0 seconds means zero lost transactions. RTO measures how fast your systems must be back online after the crash—for example, an RTO of 5 minutes means full service restoration within 5 minutes.
In Nairobi and other cities, competing ISPs often lease fibers in the same underground duct or enter your building through the same street trench. Our Physical Diversity Audit traces the actual physical manhole and duct coordinates to ensure true North/South path separation.
Yes! Using Cloud-Native Pilot-Light DR (via Veeam, Zerto, or Azure Site Recovery), your data replicates continuously to inexpensive cloud object/block storage, and full compute VMs are only spun up and billed during an actual failover or quarterly drill—cutting DR costs by 65%.
Yes. CBK Prudential Guidelines and ISO 22301 mandate documented, live Disaster Recovery failover tests at least semi-annually or annually, complete with board-signed RTO/RPO audit evidence.
An immutable backup uses hardware/cryptographic Write-Once-Read-Many (WORM) object locking with an independent time clock. Even if an attacker compromises your root Domain Admin and backup console credentials, the storage array mathematically refuses to delete or encrypt the backup files.
Yes. Our 5-day SPOF & Resilience Audit traces every electrical, cooling, network, storage, and software dependency and delivers a prioritized remediation blueprint.
Submit your current RPO/RTO targets in the form below.
Need direct architectural clarification?
Call +254 799 000 595
DIRECT DISPATCH DESK
Request Technical Briefing or Tender BOQ Quote
Routed directly to our Senior Engineering Board at 12th Floor, Mirage Towers, Westlands, Nairobi.