In this episode of Across the Tech Pond, David Marshall, Antony Savvas, and I are joined by Bill Andrews, President and CEO of ExaGrid, and Marc Crespi, Director of Business Development and Chief Evangelist. Together, we examine why backup storage has become an increasingly important part of enterprise cybersecurity, recovery planning, and long-term infrastructure economics.
Bill begins by explaining the differences between primary storage, inline deduplication appliances, and ExaGrid’s tiered backup storage architecture. His argument is that storage designed for production applications may become expensive as retention requirements and data volumes grow. Inline deduplication can reduce storage consumption, but it may also affect backup and recovery performance because data is processed before being written and rehydrated during restoration.
ExaGrid’s approach combines a front-end landing zone for recent backups with a separate repository tier. According to Bill, this design aims to preserve backup and restore speeds while reducing the storage required for longer retention periods. The repository tier is also intended to remain inaccessible to attackers operating through the production network.
We discuss ExaGrid’s expanded support for Cohesity DataProtect and why compatibility across over 25 backup applications matters. Subscription licensing allows businesses to reconsider their backup software every few years, but changing the application should not automatically require another storage replacement. Supporting platforms including Veeam, Commvault, Rubrik, Cohesity, NetBackup, and IBM Spectrum Protect can give customers greater flexibility.
The conversation also examines consolidation across the backup market and the growing relationship between data protection and cybersecurity. Backup applications already have agents operating across business systems, while ransomware groups increasingly target recovery data before encrypting production environments.
Marc describes how ExaGrid applies AI to abnormal deletion detection. The system learns the patterns created by routine backup retention and can respond when deletion activity falls outside normal behavior. Bill gives the example of a regular 30-terabyte purge suddenly becoming a request to delete 100 terabytes. Such a change could provide an early indication that an attacker is attempting to remove the company’s recovery options.
We also discuss how much autonomy customers are prepared to give AI. Marc says organizations are interested in practical, governed applications but remain cautious about systems taking independent action. Detection and temporary protection may earn confidence faster than unrestricted automation.
Bill and Marc finish with practical purchasing advice. Companies should map their workloads, recovery requirements, data growth, security needs, support expectations, and performance demands before selecting backup technology. Marc also urges buyers to assess the complete economic life of the platform rather than concentrating exclusively on the purchase price.
Does your organization evaluate backup storage as part of its cybersecurity and business continuity strategy, or does it remain an afterthought behind the backup application? Listen to the conversation and share your thoughts with us.