Best practice is to use four LUNs for the DATA disk group to allow the host to use simultaneous threads. The REDO and FRA disk groups should use two LUNs each.

References: https://community.emc.com/community/connect/everything_oracle/blog/2016/03/31/best-practises-of-vblock-on-xtremio-with-oracle-databases-part-i

XtremIO with VPLEX provides for both synchronous and asynchronous replication.

Note: The engine of RecoverPoint is the RecoverPoint Appliance (RPA).

All XtremIO's enterprise features (including Inline Data Reduction, snapshots, XDP, HA, etc.) have been developed as part of the scale-out architecture. All data and metadata are evenly distributed across the entire cluster. I/Os are admitted to the array via all the host ports, utilizing SAN zones and multi-pathing.

References: Introduction to the EMC XtremIO STORAGE ARRAY (April 2015), page 39

The VAAI XCOPY I/O size of 256 kB gives the best performance. 4 MB is the default value.

References: https://www.emc.com/collateral/white-papers/h14279-wp-vmware-horizon-xtremio-design-considerations.pdf, page 57

Choose an EMC XtremIO system and scale out linearly by adding more XtremIO X-Bricks.

References: https://store.emc.com/en-us/Product-Family/EMC-XtremIO-Products/EMC-XtremIO-All-Flash-Scale-Out-Arra y /p/EMC-XtremIO-Flash-Scale-Out

In order to use the XtremIO VSS provider it must be installed on the server where we want to do an application consistent snapshot.

References: http://muegge.com/blog/tag/xtremio/

XtremIO storage is natively thin provisioned, using a small internal block size. This provides fine-grained resolution for the thin provisioned space.

All volumes in the system are thin provisioned, meaning that the system consumes capacity only when it is actually needed. XtremIO determines where to place the unique data blocks physically inside the cluster after it calculates their fingerprint IDs. Therefore, it never pre-allocates or thick-provisions storage space before writing.

References: Introduction to the EMC XtremIO STORAGE ARRAY (April 2015), page 22

References: https://docs.openstack.org/juno/config-reference/content/XtremIO-cinder-driver.html#xtremio-management-ip

Noting the inefficiencies in VMware’s NMP driver, EMC developed a set of drivers specifically designed to overcome these limitations and improve the performance and reliability of the data passing between an array and a server. EMC developed the PowerPath family of products optimized specifically for Linux, Microsoft Windows, and UNIX Operating Systems as well as PowerPath/VE for VMware vSphere and Microsoft Hyper-V hypervisors.

PowerPath is installed on hosts to provide path failover, load balancing and performance optimization VPLEX engines (or directly to the XtremIO array if VPLEX is not used).

Note: VMware, with the cooperation of its storage partners, developed a Native Multipathing Plug-in (NMP). VMware NMP was designed to distribute the load over all the available paths and provide failover protection in the case of path, port or HBA failure, but it has not been fully optimized to work with the controllers in a storage systems. VMware’s NMP Round Robin policy does not have the intelligence that PowerPath has as PowerPath uses testing and diagnostics to continually monitor an environment to determine the optimal path for queuing requests and will adapt to current conditions.

References: https://www.emc.com/collateral/analyst-reports/emc-taneja-group-powerpath-tb.pdf