Data protection for modern workloads: What’s new in Cohesity DataProtect

Enterprise infrastructure is evolving rapidly, and data protection has to keep pace. Cloud databases, AI pipelines, containerized applications, and SaaS platforms now sit alongside traditional on-premises workloads, yet most protection solutions were never designed for this reality. The result is a patchwork of point tools, coverage gaps, and an expanding attack surface.  

Our latest Cohesity DataProtect release broadens protection for modern workloads and extends governance, threat detection, and recovery to AI-native platforms. Highlights of the release include new enhancements to help ensure that all critical data is protected, secured, and recoverable. 

AI workloads and vector databases

AI adoption in the enterprise is no longer a future state as organizations accelerate RAG-based AI application development. Vector databases, embedding stores, and managed ML platforms are now production infrastructure. The data they contain is as business-critical as anything in a traditional data warehouse.

The latest DataProtect release brings Pinecone vector databases and Google Vertex AI under the same governance, threat detection, and recovery umbrella as the rest of your environment.

More cloud workloads, including Oracle and secure AWS environments 

As enterprises diversify their cloud strategies, DataProtect support remains consistent. Most enterprises aren’t running on a single cloud, and protecting workloads consistently across all of them is imperative. DataProtect can now be deployed natively on Oracle Cloud Infrastructure (OCI), joining AWS, Microsoft Azure, and Google Cloud as a fully supported platform. For enterprises already invested in OCI, this means consistent, capable protection without compromise. 

For organizations operating in highly secure or classified environments, DataProtect is now qualified for deployment in AWS Secret East—meeting the stringent requirements of public sector and defense customers. 

Cloud databases and PaaS: Protection at the speed of adoption

Managed cloud databases have gone from experimental to foundational in just a few years. DataProtect continues to expand its coverage for native cloud databases, with key new notable capabilities: 

• Amazon DocumentDB: Cohesity now supports the protection of Amazon DocumentDB at the database-level by ingesting data into the DataProtect cluster on AWS, bringing one of the most widely used document databases under consistent enterprise-grade protection 
• Azure SQL Database with Change Data Capture (CDC): CDC support for Azure SQL Database backups enables efficient incremental ingestion 
• Azure SQL Managed Instance without CDC: Full backups using the native backup method are now supported for environments where CDC isn't enabled, improving backup speed and efficiency. 

Full and incremental ingestion is supported across each, enabling consistent, efficient, and cost-effective protection that scales with their cloud database footprint. 

Deeper Microsoft 365 protection

Microsoft 365 remains the backbone of enterprise collaboration and a high-value target. DataProtect delivers deeper protection across the full suite, including backup and recovery for OneNote files, System Document Libraries, and Teams templates. Restored files for SharePoint, OneDrive, Teams, and Groups now include key metadata (Created/modified by, Created/modified date) preserving full fidelity. Mailbox recovery now runs per-mailbox in parallel, with separate PST files and logs per mailbox, so large restores finish faster and with greater precision. 

Stay secure during, before, and after an attack

Speed of recovery matters, but so does the integrity of what you're restoring.  

Windows and Linux backup agents now use FIPS-certified OpenSSL 3, enabling compliant deployments for public sector and other regulated industries. SSH network isolation and MFA for quorum workflows add layers of protection, making it significantly harder for any single actor to tamper with backup data. Backups remain stored in immutable snapshots that cannot be accessed, deleted, or encrypted from outside the Cohesity cluster or offsite replicas. 

Recover faster, with confidence 

DataProtect delivers recovery at scale and with precision. Instant Mass Restore can bring back thousands of VMs at once, including new support for Red Hat OpenShift Virtualization, so entire environments can be restored in minutes after an outage. Azure workloads now support true file- and folder-level recovery, letting teams restore exactly what was lost without touching the rest of the VM, which is ideal for high-volume, time-sensitive help desk requests. 

Across on-prem, cloud, and SaaS, everything is centrally managed through Cohesity Helios. Granular per-workload reporting shows what's protected, anything that’s failing, and where capacity is going, while an AI-powered operational assistant reduces day-to-day admin effort. Helios also now surfaces job-level consumption data, making it even easier to track usage and justify spend. 

Learn more

Whatever cloud your data lives in, infrastructure you run, or AI tools you employ—protection, visibility, and recovery remain comprehensive and robust with DataProtect.  

To learn more about what's new, visit cohesity.com/platform/dataprotect