Single Scalable, Reliable and Performance Data Platform
HPE Ezmeral Data Fabric features synchronous scaling of files and objects that reduces the need to maintain separate technology stacks and file systems. Multiple data types are stored in the persistent data store.
Single, global namespace simplifies application or interface access to data, regardless of where it resides.
Policy-based replication delivers enterprise-level high availability by reducing single points of failure across multiple nodes, clusters, and racks.
Built-in analytics reduces data copying across multiple systems with in-place data analysis, model refinement, and fast test modifications across edge to cloud.
Automatic load balancing enables multiple applications to access common datasets without congestion and hotspots.
Fine-tuned for Analytics by Providing Multiple APIs like Apache Spark SQL, Apache Drill and Hive
HPE Ezmeral Data Fabric's data store is tuned for advanced analytics, allowing customers to orchestrate both apps and data from a single control plane to increase productivity.
Unified, secure access to a single data store delivers a single source of truth increasing data engineering and data science productivity. Tap into data repositories that are not being managed by HPE Ezmeral Data Fabric to import additional data for experimentation and model training.
Auto-tier policies balance cost and performance by automatically moving data from high-performance to low-cost storage.
Integrated S3 API is combined with HPE Ezmeral Data Fabric technology to deliver performance and scalability of all object sizes.
Unified Management for Every Data Type
HPE Ezmeral Data Fabric utilizes platform-level automation and orchestration with policies to define security, access control, snapshots, mirrors, and replication of volumes, tables, objects, and streams.
Supports multi-protocol access via integrated S3, NFS, POSIX, REST, HDFS, and container storage interface (CSI) APIs to simplify coding and data analytic model architectures.
Point-in-time snapshots create historical records for test/dev experiments as well as process and model validations by data scientists and data engineers.
Delivers flexibility to data science teams to restore entire snapshot or individual files within the snapshot.
Integrated platform for CDC and IoT data coming into Event Streams and then persisting it.
HPE Ezmeral Data Fabric supports near real-time data ingestion into Event Streams using CDC or IoT data and then allowing transformation and persistence in Files, Objects and/or NoSQL databases.