When you are working with massive datasets spread across locations and storage systems, slowdowns aren’t just frustrating, they are costly. Data silos, performance bottlenecks, and manual handoffs can disrupt critical work like HPC simulations, AI training, and high-resolution media rendering. As environments scale, so do the risks: missed deadlines, spiraling costs, and stalled innovation.
DataCore Nexus removes these barriers. It combines the raw speed of Pixstor high-performance parallel file system with the intelligent automation and global data mobility of Ngenea orchestration engine. The result is a unified platform that accelerates demanding workflows, automates data movement, and puts you in control of your data, no matter the scale or complexity.
- Ultra-fast, parallel file storage built for high-throughput, low-latency access
- Consolidate siloed storage into a unified, software-defined platform
- Accelerate data-intensive compute, scientific, AI, and research workflows
- Automate data movement across NVMe, cloud, object, tape, etc.
- Enable global collaboration and seamless remote access
- Protect and replicate data efficiently without overhead
- Scale without compromise: on-premises, in the cloud, or hybrid
Cas d’usage
Drive HPC and research efficiency with scalable I/O and smart data movement
Accelerate AI/ML workflows with high-speed access and smart data flow
Streamline media pipelines from ingest to archive and global delivery
Enable global collaboration with consistent access, remote workflows, and smart recall across sites
Support hybrid and cloud workflows with automated tiering, caching, and policy-based data placement
Simplify lifecycle management by automating data movement based on usage, cost, and retention goals
Nexus is built for environments where data access speed makes or breaks results. With line-rate performance at petabyte scale, it ensures critical workloads stay on track, no matter how big or complex the dataset (e.g., HPC, AI, media pipelines).
- Parallel file system with POSIX compliance for extreme concurrency
- Predictable low-latency performance for high-throughput workflows
- Scales seamlessly to handle billions of files and mixed workloads
- Tuned for performance across NVMe, SSD, and HDD tiers
- Delivers up to 180 GB/s read performance in 4U form factor
Access your data, your way. Nexus provides a single global namespace that spans locations and storage types, with native support for simultaneous file access over multiple protocols.
- Unified view across NVMe, SSD, HDD, object, cloud, and tape
- Multi-protocol support: NFS, SMB, S3, POSIX, RESTful API access
- Supports NVMe-oF and GPU Direct for ultra-fast, low-latency data paths from storage to compute and GPU nodes
- Logical segmentation via filesets for access control
- Namespace isolation for project and tenant separation
- Seamless path retention during data movement
Nexus offers lightweight, scalable snapshot and cloning capabilities that enable rapid recovery, rollback, and staging without disrupting active workflows or requiring extra tools.
- Copy-on-write snapshots with minimal overhead
- Writable clones for test, staging, or temporary use
- Mountable snapshots for instant access or DR
- Versioning support for historical recovery
Find what you need, fast. Nexus includes built-in metadata services for tagging, indexing, and searching across live, archived, or remote datasets – all without rehydrating data first.
- Full-text and attribute-based file search
- Content-aware indexing (EXIF, AI tags, owner, size, etc.)
- Custom tagging using extended attributes (xattrs)
- Search across all storage tiers, including cloud and tape
- Accessible via web UI or programmable API
Automate data movement between performance and capacity layers with customizable policies. Nexus ensures data is always in the right place – fast when needed, cost-efficient when not.
- Rule-driven tiering by age, usage, location, tags, and more
- Seamless migration across NVMe, SSD, HDD, object, tape
- Retains file paths, metadata, and permissions during movement
- Supports hydration (recall) on demand or via schedule
- Integrates with job schedulers for compute-aware staging
Nexus is built with strong multi-tenant security, offering granular control over user access, data isolation, and compliance even in large-scale shared environments.
- Role-based access control (RBAC) across users and groups
- Active Directory, LDAP, OAuth2, and OKTA integration
- Secure project- and tenant-level segmentation
- Snapshot immutability and retention enforcement for compliance
- Encryption in transit (TLS/SSH) and optionally at rest
Protect your data across locations with intelligent, efficient replication. Nexus uses snapshot differencing for fast sync and supports flexible recovery without duplicating entire datasets.
- Site-to-site replication with minimal data movement
- Snapshot-based sync for fast incremental updates
- Mountable DR sets with full file access
- Supports point-in-time recovery and rollback
- Self-service restore via standard file protocols
Gain control and visibility with built-in tools for monitoring, automation, and lifecycle governance. Nexus adapts to your environment and integrates cleanly with orchestration tools.
- Real-time metrics on performance and usage
- Slack/email alerting, API and Python SDK access
- File-based automation (e.g., ingest triggers workflow)
- SaltStack-driven system configuration and policy enforcement
- Integrates with job schedulers (such as Slurm) and automation frameworks
File Usage Analytics
Observability & Monitoring
Policy-based Data Management
Avantages
Accelerate Time-to-Results
- Slash job start delays with instant access to datasets
- Pre-stage data intelligently for compute workflows
- Eliminate storage-related bottlenecks in critical pipelines
Work Smarter with Automation
- Set and forget lifecycle policies based on real usage
- Automatically move data between fast and deep storage
- Reduce manual file management and guesswork
Find What You Need, Fast
- No more hunting across systems: unified view of all files
- Search by project, user, or tags even if data is archived
- Actionable search: copy, move, archive in one step
Lower Costs Without Compromise
- Keep expensive storage focused on active work
- Maximize ROI by using commodity hardware
- Avoid cloud surprises with built-in cost-aware controls
Enable Global Collaboration
- Remote teams work as one through unified namespace
- Stub access ensures files are always discoverable
- Files are recalled on demand, exactly when and where they are needed
Plug Into Any Workflow
- Works with your existing apps, workflows, and infrastructure
- Integrates with job schedulers and pipelines
- No vendor lock-in: data stays in your control
Architecture
DataCore Nexus (Pixstor & Ngenea) Reference Architecture for HPC/AI Environments