HPE Ezmeral Data Fabric
HPE Ezmeral Data Fabric’s high performance, all software platform delivers the right data, at the right time, to the right application. It allows developers and scientists to use familiar tools for data intensive application development across core, edge, container and IoT environments via a wide range of APIs and deployment mechanisms. A global namespace makes it possible for multiple protocols, applications and users/teams to access the same data set without impacting multi-tenancy or security policies. HPE Ezmeral Data Fabric is perfect for:
Global 2000+ companies with data modernization initiatives
Customers that want to use software defined storage across core, edge, and IoT using commodity hardware
Accelerating time-to-data value by speeding up ingestion, curation and analysis across diverse data sources residing in core, cloud and edge environments
- File Data Store is perfect when existing solutions cannot solve requirements for high data growth, new sites (edge, cloud, IoT) and/or new applications.
- Distributed datastore for files and persistent storage for containers, cloud file migration and support for HDD and SSD.
Flexible, All-Software Platform
Single data layer consolidates systems across locations and data types into a single view that reduces management complexity and costs.
Software defined scale-out storage and management speeds up ingestion, curation and analysis of data across core, cloud and edge.
Single infrastructure allows data to be managed as a single resource across files, table and data streams.
HPE Ezmeral Data Fabric provides storage and management choice without hardware lock-in.
Cost efficient monetization of large-scale data sets.
Data clusters span geographic locations to deliver consistent file system structure to applications.
Scale-out file and object system across multiple physical and virtual clusters managed through one logical instance.
It delivers API-independent uniform access to files, tables and event streams even if data resides on multiple clusters in different locations.
Flexibility in tool and coding language choice to run legacy or modern applications on the same data layer and tenant.
Kubernetes persistent storage across all application workloads.
Resilient and Portable
Built-in mirroring simplifies data movement between and within clusters, geographic locations or secondary data centers for DR.
Ensures reliability and performance for very large clusters with self-healing capabilities without disruption of operations or system re-architecture.
Distributed metadata is a powerful design that protects against single or high-impact points of failure resulting in data loss.
To avoid congestion, the system reads from multiple replicas using multiple links or automatically moves data to avoid hotspots.
Directory, file, table and system metadata is copied as data is being replicated across a cluster.
Optimized for Scale
As data is being ingested, automatically grow storage capacity without impacting SLAs, reliability, downtime or IT intervention.
Designed to grow to very large-scale systems and still retain service levels for availability, performance and latency.
Automated storage tiering across hot, warm and cold layers optimizes utilization, cost and performance.
Linux is the registered trademark of Linus Torvalds in the U.S. and other countries. Red Hat is a registered trademark of Red Hat, Inc. in the United States and other countries. Oracle is a registered trademark of Oracle and/or its affiliates. All third-party marks are property of their respective owners.
Find what you are looking for?