Standards and Best Practices in Disaster Recovery

Abstract

The digital landscape necessitates robust Disaster Recovery (DR) capabilities, driven by the need to maintain low Recovery Time Objective (RTO) and Recovery Point Objective (RPO) targets amidst escalating cyber threats. This paper addresses the critical transition from theoretical DR standards such as ISO 22301 and NIST SP 800-34 to practical, secure, and cost-effective operational solutions. While the literature provides extensive normative guidance, a gap persists in the unified implementation and performance validation of complementary open-source tools. The research employs an applied and comparative experimental methodology, testing the primary hypothesis that an integrated solution built on Red Hat Enterprise Linux (RHEL 9.5), leveraging Relax-and-Recover (ReaR) for Bare-Metal Recovery (BMR) and BorgBackup (BORG) for archiving, yields superior overall resilience. The system architecture utilizes an Ansible-automated multi-node environment, secured via SSH/GPG, to ensure consistent configuration and data protection. Main results confirm that the tools are strategically complementary: ReaR provides the predictable recovery path essential for a low RTO, while BORG achieves superior RPO and storage efficiency through block-level deduplication (over 85% efficiency recorded) and client-side authenticated encryption. The append-only mode configuration of BORG repository directly provides ransomware immutability. The paper's contribution lies in validating this integrated, open-source framework as a secure, measurable, and highly effective model for organizations seeking to achieve compliance and strengthen their operational resilience against catastrophic failure.