Table of contents
- What’s new in Hystax Acura 4.5 — at a glance
- Direct2Target — VM migration without target platform API dependency
- Backup 2.0 — a redesigned storage layer
- PostgreSQL Backup
- S3 Object Lock — Immutable Restore Points
- New target platforms
- Performance improvements for oVirt-based platforms
- Audit security management and security updates
Hystax Acura 4.5 is a significant release that addresses three major areas: a new approach to VM migration that removes dependency on target platform APIs, a fundamentally redesigned storage layer for backup, and a meaningful expansion of supported target platforms. Combined with security management improvements and stability fixes across the board, this release advances Acura on multiple fronts.
Below, we walk through the key changes and what they mean in practice.
What’s new in Hystax Acura 4.5 — at a glance
Key highlights of this release include:
- Direct2Target: VM migration and replication without dependency on target platform APIs
- Backup 2.0: a new storage layer with storage-wide deduplication, encryption, configurable compression, and improved maintenance
- PostgreSQL Backup: a dedicated file-level agent with native PostgreSQL backup/restore flow integration
- S3 Object Lock support: immutable restore points protected against deletion or modification
- New target platforms: STACKIT, CloudSigma, Red Hat OpenShift Virtualization, Oracle Linux KVM, and Oracle Linux Virtualization Manager
- Performance improvements for oVirt-based platforms
- Audit security management: end-to-end audit event tracking across platform services
- General security updates
Below, we take a closer look at each of these improvements.
Direct2Target — VM migration without target platform API dependency
VM migration has traditionally required Acura to interact with the target platform’s management APIs — to provision VMs, attach disks, and manage recovery. This works well for platforms where those APIs are accessible and well-defined. But in practice, many organizations encounter environments where this assumption breaks down: private clouds with limited API exposure, air-gapped infrastructure, edge deployments, or platforms that support disk-level write access but lack a standard management interface. In these scenarios, the API dependency becomes a blocker, forcing teams to either avoid the platform entirely or work around it with custom integrations.
Direct2Target introduces a new migration approach in which Acura replicates VM data and performs recovery entirely at the disk level, bypassing target platform management APIs. A dedicated D2T agent is deployed on the target side — downloadable directly from the Acura control panel and identifiable by IP address in the UI. The agent handles replication receipt, disk layout, and recovery without requiring the target platform to expose an API. It supports both BIOS and EFI boot configurations, uses NetworkManager for network management (consistent with other Acura chassis-based agents), and integrates with the standard Acura replication and recovery workflows.
Direct2Target makes the target platform’s API availability irrelevant to the migration process. Any environment that supports disk-level write access can now serve as a migration or recovery target, regardless of whether it exposes a management API. This expands the range of platforms Acura can work with and removes a class of blockers that previously required workarounds or prevented migrations entirely. For organizations operating across diverse or non-standard infrastructure, it means consistent tooling and workflows, even where platform-level API access is restricted or absent.
Backup 2.0 — a redesigned storage layer
As backup environments grow — more machines, more restore points, longer retention periods — the underlying storage layer becomes a determining factor in both cost and reliability. Traditional block-level backup approaches store data per device without cross-machine awareness, which means duplicate data (common across similarly provisioned VMs) is stored repeatedly. Storage costs scale linearly, maintenance becomes more complex, and there is no built-in mechanism to protect backup data at rest.
Backup 2.0 is a new storage layer built on different architectural foundations from the previous backup system. Its core capabilities in this release:
- Storage-wide deduplication: the deduplication engine operates across the entire storage pool rather than per device. Identical data blocks across different machines and restore points are stored only once, reducing the total storage footprint in environments with similarly or identically provisioned VMs.
- Encryption: backup data is encrypted at the storage layer, providing data-at-rest protection without requiring external key management infrastructure.
- Configurable compression algorithms: organizations can select from multiple compression algorithms to balance throughput, compression ratio, and CPU overhead based on workload profile.
- Improved storage maintenance and cleanup: the new layer includes updated routines for managing restore points, cleaning up orphaned data, and keeping storage utilization predictable over time.
The storage format is compatible with the Kopia open-source backup engine, providing a well-understood, community-vetted foundation for deduplication logic and storage layout.
For environments with large numbers of similarly provisioned VMs — common in DevOps pipelines, VDI deployments, or templated cloud infrastructure — storage-wide deduplication can substantially reduce storage costs relative to per-device approaches. Encryption addresses data protection requirements without adding operational complexity. Configurable compression gives infrastructure teams a tuning lever for different workload types. Together, these capabilities make the backup storage layer more cost-efficient, more secure, and more maintainable at scale.
your DR, backup, and migration strategy
Leave your email to book a personalized demo.
Thank you for your request!
We will be in touch soon.
We respect your privacy. See our Privacy Policy. You can unsubscribe at any time.
PostgreSQL Backup
Backing up a PostgreSQL database by taking a VM-level snapshot captures the disk state but does not ensure the database’s internal consistency. A snapshot taken while transactions are in flight may produce a backup that is technically complete at the block level but inconsistent at the database level — requiring recovery steps that add time and risk. Database administrators working with PostgreSQL typically need backup tooling that understands database semantics, not just disk state.
Hystax Acura 4.5 adds a dedicated PostgreSQL backup agent — a specialized file-level backup component that integrates with PostgreSQL-native backup and restore flows. Rather than capturing a disk snapshot of the machine running PostgreSQL, the agent participates in the database’s own backup process, ensuring consistency at the database level. Recovery workflows align with PostgreSQL administrator practices and use the database’s native restore mechanisms.
Organizations running PostgreSQL as a self-hosted or managed database can now use Hystax Acura to protect it with tooling that understands database-level consistency — without resorting to VM-level snapshots that require additional recovery steps. This is particularly relevant for teams that already use PostgreSQL backup conventions and want their data protection tooling to fit those workflows rather than replace them.
S3 Object Lock support — Immutable Restore Points
One of the most effective ransomware attack vectors against backup infrastructure is targeting the backup storage itself. If backup data can be deleted or overwritten — by malware, by a compromised account, or by a malicious insider — the backup ceases to be a reliable recovery option. Regulated industries increasingly require formal guarantees that backup data cannot be tampered with during its retention period, and standard access controls alone are not sufficient to meet this bar.
Hystax Acura 4.5 adds support for S3 Object Lock, enabling immutable restore points. When configured, backup data written to S3-compatible storage is locked for a defined retention period and cannot be deleted or overwritten by any means — including administrative access, automated processes, or ransomware. Key implementation details:
- Object Lock retention settings are configurable per cloud storage target
- The UI displays warnings when Object Lock is active, preventing accidental misconfiguration of retention periods
- Delete operations on machines or backups with active locks are handled gracefully with appropriate guardrails
- Legal hold is supported for scenarios requiring indefinite retention outside a fixed retention window
- Immutable backups are compatible with both block-level and file-level backup flows in Acura
Immutable restore points provide a credible guarantee that backup data will be available for recovery even if the primary infrastructure is compromised. For regulated industries — financial services, healthcare, public sector — this directly addresses compliance requirements around backup integrity and retention. For any organization concerned about ransomware resilience, it closes a significant gap in the data protection posture: a backup that cannot be deleted is meaningfully more valuable than one that can.
New target platforms
Acura 4.5 expands the list of supported migration, disaster recovery, and backup targets with four additions:
- STACKIT — the sovereign European cloud platform operated by Schwarz Group (Lidl/Kaufland), designed for GDPR-aligned deployments with EU data residency. Organizations looking for a European alternative to hyperscaler public clouds now have a direct migration and DR path via Acura.
- CloudSigma — a Swiss IaaS provider offering highly customizable cloud infrastructure. CloudSigma’s model is particularly relevant for organizations that need fine-grained control over compute resources or prefer a European provider outside the major cloud ecosystems.
- Red Hat OpenShift Virtualization — an included feature of Red Hat OpenShift that helps organizations migrate, run, and manage traditional VMs alongside modern workloads on a unified, hybrid cloud application platform. Acura now supports this as a migration and DR target, building on the earlier support announcement and including compatibility and stability improvements in this release.
- Oracle Linux KVM and Oracle Linux Virtualization Manager — extending Acura’s support for Oracle’s open-source virtualization stack, relevant for organizations running Oracle Linux-based infrastructure who need migration or DR tooling integrated with their existing environment.
These additions reflect the growing demand for migration and data-protection paths to alternative cloud providers, including sovereign and compliance-focused infrastructure options in Europe.
Performance improvements for oVirt-based platforms
Acura 4.5 includes targeted performance improvements for replication on oVirt and oVirt-derived platforms. The changes address specific bottlenecks in the data transfer path, improving throughput and reducing replication time for large VMs or high-volume replication workloads. Additional improvements include better cloud agent load balancing across available agents for VMware and oVirt sources, and a configurable limit on the number of concurrent replications to prevent resource exhaustion under high load.
Audit security management and security updates
In multi-tenant and security-sensitive environments, operational visibility is not just a convenience — it is a compliance requirement. Understanding which actions were taken, when, and by which components is essential for incident investigation and for satisfying audit obligations. Without a structured audit trail, these questions are difficult to answer reliably from logs alone.
Acura 4.5 introduces a new audit event framework that captures and propagates security-relevant events across the platform:
- Audit events are collected from the REST API layer and propagated through core platform services
- A trace identifier is attached to device state changes and passed through to agents, enabling end-to-end tracing of operations across distributed components
- Audit event data is surfaced via the Grafana-based observability stack included in the platform
The release also includes a set of general security updates addressing internal dependency versions and platform hardening items.
The audit event framework provides a structured, queryable record of what happened across the platform — useful for security incident investigation, compliance reporting, and operational troubleshooting. For organizations subject to regulatory audit requirements, this closes a gap that previously required external tooling or manual log correlation. For security teams, end-to-end traceability across distributed components makes it significantly easier to reconstruct the sequence of events during an incident.
To review all version updates, visit the full Hystax Acura release notes or contact us.
