Claritum Architecture Security and Disaster Recovery
- Kasia Czwarkiel (Unlicensed)
Architecture
The platform runs on Red Hat Enterprise and SELinux servers with MySql databases and PHP.
Multiple virtualised servers are used to ensure that there is no single-point-of-failure and each server comprises multiply-redundant components.
The configuration is multi-tiered, with separate web servers, database servers and file servers - each cluster performing a single task to ensure rapid scalability if required.
Proactive monitoring is also in place to alert us to any system performance issues. We dynamically scale resources to meet demand.
Each of our customers databases runs in its own instance so we can ensure complete separation of data.
Data centres
Networks
All data transferred between servers inside a single region is routed over a private network. All transfers between the data center and clients (via the Claritum web-based application, via direct connections to the data warehouse or for Claritum's internal requirements) use industry standard 2048-bit encryption.
Backups
All databases are mirrored to a hot failover and are backed up nightly. Incremental daily snapshots are kept encrypted for a minimum of six months. The database transaction logs are similarly stored in encrypted backups for one month, ensuring we can perform point-in-time recovery for forensics or disaster recovery.
User-contributed content are stored securely in Cloud Files, with redundancy and secure backup provided by Rackspace.
These backup processes also proactively generate alerts in the event of any failure