Global high availability architecture 

Advanced high availability architecture

Annex Cloud’s cloud-based infrastructure is designed to be highly available. All servers and network devices have redundant components and multiple network paths to avoid single points of failure.

Production application servers are load balanced within each region. Production database servers are replicated in near-real time to a peer data center within the same geographic region in Europe, North America, and Asia-Pacific.

In the event of the failure of one or more infrastructure components, service is restored by transferring the operation of an application associated with the failed components to the peer data center.

Global data center pairs

Offer bonus points for purchasing must-have items to ensure loyalty members look to you for their day-to-day items. Aside from promotions tied to transactions, your loyalty program provides you with the opportunity to keep customers engaged between purchases, and recognize and reward them for non-transactional behaviors, such as downloading your app, sharing on social media, referring others, and more.

Annex Cloud maintains continuous, asynchronous replication from the database in the current primary data center (read-write) to the secondary data center (read-only). To transfer application instances from a primary to a secondary data center, Annex Cloud designates the secondary to be the primary and the primary to be the secondary.

• Multiple web farms (each comprised of web servers, database servers, firewall, cache, etc.) in the same zone/region to share load 
• Auto-scaling with both scaling up and out on both web and database nodes
• Dynamic load balancing for both web and database nodes

Availability

Annex Cloud aims for our platform to be operational 24/7, 365 days a year. No vacation, no extended upgrade or maintenance windows, no single points of failure.

We focus on near-perfect availability with redundancy built into every layer, including redundant devices, all network and server infrastructure. There’s no downtime for upgrades. Real availability is the true measure of customer availability by looking at every incident that results in a customer outage (a Priority 1 or P1 incident).

• Redundant nodes available and ready to take over during failover within clusters 
• Real-time availability and heartbeat monitoring with failover
• 99.9% monthly uptime service commitment across all global regions

Performance

Annex Cloud’s cloud-based architecture scales to meet enterprise needs. We have hundreds of customers operating globally.

Each of our customer instances leverages our multi-tenant architecture to perform an aggregate of tens of millions of full-page transactions every month.

• Scalable caching layer (Redis) to reduce database load
• Global CDN and edge-caching to reduce web load
• Architected to allow client caching for key data, calling Annex Cloud only for updates/real-time needs

Database backup and recovery

While Advanced High Availability (AHA) is the primary means to recover data and restore service in the case of a disruption, in certain cases it’s desirable to use Annex Cloud’s more traditional data backup and recovery mechanism. It works in concert with AHA and acts as a secondary recovery mechanism.

The backup cycle consists of full backups of 30 days and the past seven days of daily differential backups every day. All backups are written to disk, no tapes, and no backups are stored in different regions. All controls that apply to live customer data also apply to backups. If data is encrypted in the live database, it will also be encrypted in the backups.

Regular automated tests are run to ensure the quality of backups. Any failures are reported for remediation. Real-time off-zone and region data backups are performed.