XIOS 4.0.10 And XMS 4.2.0 Are Here, Here’s What’s New


I’m so proud to finally announce the GA of XIOS 4.0.10/ XMS 4.2, this release contains couple of features that so many of you, our customers have been asking for so let’s highlight them and then deep a little bit deeper to each one of them

ŸXtremIO PowerShell support
ŸSMI-S support
ŸServiceability enhancements
ŸREST API Improvements
ŸVSS enhancements
ŸXMS Simulator
ŸWebUI Tech Preview
Ÿthe usual bug fixes.

XtremIO PowerShell support

  • Supported with XMS 4.2.0
    PowerShell Supported versions: 4.0 and 5.0
    Based on XtremIO REST API version 2.0
    Supports all storage management commands


    XtremIOPowerShell.msi package – available on support page
    Verify PowerShell ISE version 4 and above is installed
    Installation package imports the module to PowerShell path

Connecting To The XMS

Connect to single cluster or all clusters manages by the XMS

Powershell Commands

Supports storage management REST commands
XtremIO command structure:
Supported actions:
Example: Get-XtremVolumes


Ability to list a subset of attributes (-Properties)
Filtering logic support (-Filters)
-CLImode: avoid user confirmation in session level (for scripting)
-Confirm: user confirmation for single command
-ShowRest: returns the command in json format
-Full: list all object attributes





SMI-S Integration

What is an SMI-S Provider?

SNIA developed standard intended to facilitate the management of storage devices from multiple vendors in storage area networks.
Enables broad interoperable management of heterogeneous storage vendor systems
Multiple ‘Profiles’ can be implemented
We have implemented the profiles required for Microsoft SCVMM and Microsoft Azure platforms
All SCVMM operations can be done in the GUI or through CMDlets
More profiles would be implemented in future based on field requirements and roadmap features


SMI-S Provider implemented directly on the XMS
No external component needs to be installed
Better and consistent performance guaranteed
Array operations possible
Create/delete a volume
Create an XtremIO Virtual Copy – XVC (aka Snapshot)
Mount the volume on a host
The entire array is considered as a single pool
Internally uses RestAPI calls
Completely stateless and does not cache any data

ECOM needs a new Read Only user on XMS
Needs to be defined on ECOM (for non-default password usage)
Same user to be defined on ECOM and XMS for adding
Provide the same credentials in SCVMM
LDAP users also supported

The following profiles have been implemented:
Masking and Mapping
Disk Drive Lite
FC Target Ports
iSCSI Target Ports
Physical Package
Multiple Computer System
Block Services Package
Thin Provisioning
Replication Services (Snapshots)



Fabric Tab -> Add
Resources -> Storage
Specify the “Run As” account as defined in ECOM
and XMS
Go to the ‘Jobs’ tab to see operation status
‘Providers’ option will show the XMS information and current status


Create Logical Unit’
allows to create new volumes
Right click on volume name to delete

Right Click on Host -> Properties

Allow active clusters to send SNMP keep alive trap to SNMP manager
Can be sent between 5 min to 24 hours , customer configurable
Working via CLI require two commands
1. Enable the feature and frequency on XMS level:
modify-snmp-notifier enable-heartbeat heartbeat-frequency=15
2. Enable the feature on cluster level (default is enabled for all):
modify-clusters-parameters enable-heartbeat


ŸGUI and CLI now support the option to export and
XMS configuration for back-up and restoration
ŸThe following configuration elements are exported
XMS: User Account, Ldap Config, Syr Notifier, Syslog Notifier, Snmp Notifier, Email Notifier, Alert Definition
Cluster: Cluster Name, X-Brick Name, SC Name, Target Ports, iSCSI Portal & Routes, IG, Initiator, Volume, CG, LUN Mapping, Tag, Scheduler

Snapshot Enhancements
Native VSS support for application aware snapshots
Snapshot Enhancements
The new VSS supports working inside the VM guest using RDM volumes.

WebUI Technical Preview

Yea..that’s the one you have all been waiting for but couple of disclaimers, it’s a technical preview which means we ask you to test and provide feedback for, it’s not the final Web UI and It’s likely to change before GA and again, the reason we are releasing it is that you can contact us and let us know your opinion about it, good, bad, ugly, it’s all good! Please note that the classic Java is still working and provides the full functionality, so it’s

100% pure HTML5 (no Java) !!!
Just enter XMS WebUI URL
Enter your standard XMS User credentials
Nothing to install

In Multi-cluster setup: Multi-Cluster Overview
In Single cluster setup: Single-Cluster Overview
To access WebUI Homepage: Click on WebUI Logo

Single cluster homepage


2 main navigation
Menu bar
Context selector

Only object types supported
by the selected menu items will appear in the Context Selector
Filtering capabilities:
Direct: Text, selected properties, tags
Indirect: Filter based on relationship to other objects






For each object type in the Context selector there is a list of supported reports
ŸAll reports support single/multiple object reports, for example:

Troubleshoot an object with all available data:
One-click navigation between pre-defined reports

Track capacity and data savings over time

Track endurance and SSD utilization

Notifications Events

Events screen

ŸDrill down to Critical/Major Alerts from Cluster Overview

Storage Configuration & mapping

Configuration Screen
Create/delete, perform all actions on selected objects.

Storage configuration & mappings
Configuration Screen
No context selector
Indirect cluster filtering

Mappings Screen
Map between selected Volumes/CGs/Snapsets and selected Initiator Groups (many-to-many object mappings supported)

Main provisioning flow

(1)Create Volumes (2)Create Initiator Groups (3)Map

3. Map

Add volumes to Consistency Group

Local Protection

Create 1-time local protection or define local protection policy

Refresh copy


  • Export configuration/inventory object data

Hardware view


VMware horizon 7.0.1 / Instant Clones / App Volumes 2.11 / Windows 10 + Office 2016 On XtremIO

Wow, that was a long title for a blog post but then again, there are many moving parts in this cutting edge mixture of technologies..

Instant Clone Technology (that is, vmFork) uses rapid in-memory cloning of a running parent virtual machine, and copy-on-write to rapidly deploy the virtual machines.

Copy-on-write, or COW, is an optimization strategy. This means that whenever a task attempts to make a change to the shared information, it should first create a separate (private) copy of that information to prevent its changes from becoming visible to all the other tasks.

A running parent virtual machine is brought to a quiescent, or quiet, state and then forked (when a piece of software or other work is split into two branches or variations of development), and the resulting clones are then customized (receiving unique MAC addresses, UUIDs, and other information) and powered on. These clones share the disk and memory of the parent for Reads and are always ready for login instantly. After the clone is created, Writes are placed in delta disks. Both memory and disk are copy-on-write, so if a new clone modifies bits of its memory or disk, a separate copy is made for that virtual machine, thus preserving security and isolation between virtual machines.

Instant clones are installed as part for the Horizon Broker. There is no option to select to install, it just installs with the Horizon 7 broker.

The instant clone engine is a Java app that runs on the Tomcat server.

API calls are made to VMFork on vSphere.

The Master Image has the Horizon Agent installed with the Instant Clone option selected. This has the code for instant clones and the instant clone customization engine.

Any broker is capable of running the NGVC engine, but only one of them should be running NGVC. We use the existing Federated Tasks to manage this. If the broker running the NGVC were to crash/shutdown, then the task would move to another broker in the cluster. In flight operations can error out in that scenario, but newer cloning operations should work fine.

Stateless because all configurations information is stored in vCenter. No multiple database scenario – Composer database / LDAP / vCenter to get out of sync. One source of truth.

Master VM

  • Golden Image”
  • Follow build Guidelines in KB 2007319 – See Hotfix for Win7 in Step 10
  • Optimizations – OS Optimization Tool
  • Horizon Agent with Instant Clones
  • Other Agents – AppVolumes, FlexEngine
  • Snapshot(s)


  • Linked clone of Master VM – Based on chosen snapshot
  • Low disk space Usage
  • Named as <cp-template-GUID> in vCenter in <ClonePrepInternalTemplateFolder>
  • Defaults to same datastore as Master VM
  • Linked to Master VM.


  • Full Clone of Template – Thin provisioned – Less total disk space than Master VM
  • Has a Digest
  • Placed on selected datasore(s) for desktops. Can be placed on a single specific datastore in case of tiered storage selection.
  • Named as <cp-replica-GUID> in vCenter in <ClonePrepReplicaVmFolder>
  • Shared read disk for desktop VMs


  • Used to fork Desktop VMs
  • Placed on selected datasore(s) for desktops. One per host per datastore
  • Named as <cp-parent-GUID> in vCenter in <ClonePrepParentVmFolder>

Desktop VMs – Instant Clones

  • Placed on selected datastore(s) for desktops.
  • Named as defined in Horizon Administrator Pool Settings
  • Very small disk space usage – Can grow over time but limited by delete on logoff

Instant Clone Components in vCenter

Here’s a deeper dive session into the concepts (and more!) detailed above

ok, so now you are familiar with Instant Clones, let’s see how do they behave on XtremIO !

AppSync 3.0 Is Here, What’s New

Hi, we have just released AppSync 3.0, if you are new to AppSync, I encourage you to first read about it here


and here, https://itzikr.wordpress.com/2016/03/31/emc-appsync-2-2-sp3-is-out-another-must-upgrade-for-xtremio-customers/

in a nutshell, Appsync is our (Integrated) Copy Data Management (iCDM), IDC & Gartner predict that the bulk of data is coming from copies of the data itself and AppSync is our glue to manage it at the array level, traditionally, snapshots are always associated with degraded performance either at the array controllers level or at the volume level, even the majority of the modern All Flash Arrays can’t sustain a large numbers of snapshots because of architecture deficiencies.

This is not the case with XtremIO so the “glue” here is very important. So, what’s new with 3.0


Yep, It’s finally here, many customer requests came for this one as many customers are leveraging XtremIO + VPLEX for either a true Active/Active datacenter(s) topology or for the VPLEX availability features, Appsync now fully support VPLEX, here are the modes of supported operations

You need to be on the following versions for it to work:

VPLEX firmware version –
XtremIO firmware version 4.0.1, 4.0.2, and 4.0.4

Configuring a service plan requires selecting:
Preferred cluster selection (metro only)
Cluster 1 is chosen by default
And array preferences (when R1)
Required selection
VPLEX distributed virtual volume has storage devices on two clusters
The leg to be protected is determined by the cluster that is selected in the service plan storage options

Mount as VPLEX Virtual Volumes
– The provisioned virtual volumes are added to the storage
view of the mount host
– Teared down during unmount
Mount as native array volumes
– The mount host must be zoned to the native array where the snapshot is created


When mount host is part of an ESX Cluster, target
LUNS are presented to all nodes of the cluster

Clear this option if you do not want to perform an
ESX cluster mount


The VPLEX production virtual volume layout must be the same as it was when the copy was created
R1 and distributed devices are restored on one leg of the mirror for which the copy was created
Other leg will be rebuilt and synchronized after restore is complete
User has option to wait or run in background
AppSync does not support restore of volumes protected by RecoverPoint


Support XtremIO backend only
Support Bronze service plan only (local copies)
VPLEX virtual volumes must be mapped 1:1 to an array volume
Concatenated devices (RAID-C) are not supported
Nested devices are not supported
Exchange is not supported
Tolerant of non-XtremIO array on second leg only


You can now perform consistency check only on log files of Exchange database copy in DAG environment

below you can see a demo of AppSync 3.0 / VPLEX / XtremIO working as one.



Provides scripting opportunity during the Unmount previous copy and Unmount copy phases
Supported on all operating systems in addition to AIX
The unmount callout scripts are placed in a predefined location with a predefined name
The scripts are run with the user credentials used to register the host


Enhanced scheduling ability to repurpose and refresh a copy, including ability to
schedule repurposing jobs for 2nd Gen copies
Ability to view and edit the values of various fields in the repurpose workflow
Centralized view of all database repurposed copies


New VMware Service Plan options
Create VM snapshots for selected VMs only – ignoring newly added VMs
Can now “Exclude VMs


Supports taking a cluster copy and mounting into an alternate cluster, attaching/recovering databases as clustered resources if an alternate cluster SQL Server instance is selected during recovery
Restoring back as a cluster or non-clustered resource
SQL Server databases hosted on mount points are not supported – Mount to “Default Path” is not supported due to Microsoft restrictions.


Similar to crash consistent, but still utilizing Microsoft VSS on the file system (does not integrate with SQL)
Mount with attaching the database (recovery) on target host is not supported for non-VDI copies – all other SQL recovery options are not supported
Restoring as a file system is supported – reattaching and recovering the database is a manual process

here’s a demo about the NON-VDI integration

Please Vote for the EMC XtremIO Sessions At VMworld 2016

It’s hard to believe a year has passed to quickly since last year’s VMworld, this years, we at the EMC XtremIO have submitted many good sessions for the submission process, please vote for the ones you are interested in thanks!



Best Practices for Running Virtualized Workloads on An All-Flash Array [7574]

All-Flash Arrays (AFAs) are taking the storage industry by storm. Many customers are leveraging AFAs for virtualizing their data centers, with a majority of them using EMC XtremIO. In this session, we will examine the reasons behind strong adoption of AFAs, discuss similarities and differences between different AFA architectures, and deep dive into specific best practices for the following virtualized use cases: 1. Databases: how they can benefit from being virtualized on an AFA; 2. EUC: how VDI 1.0 started, what VDI 3.0 means and how it is applicable for an AFA; and 3. Generic workloads being migrated to AFAs. If topics like UNMAP, Queue depths etc’ rings a bell, this is the session for you!



Demystifying OpenStack Private Clouds with VMware and EMC XtremIO All-Flash Array [8330]

OpenStack Cloud deployments have long been notorious for being too complex and unpredictable, often requiring long implementation times and ongoing optimization. In this session, experts from VMware and EMC will discuss how VMware Integrated OpenStack (VIO), coupled with EMC XtremIO all-flash scale-out storage can dramatically simplify OpenStack implementations while delivering enterprise-class reliability, consistent predictable workload performance, and easy linear scalability.



Best Practices for Deploying Virtualized SQL Server on EMC XtremIO All-Flash Storage [8746]

All-flash storage opens up many new ways managing SQL Server database lifecycle, can impact your existing virtualized SQL Server design and deployment methodologies. Come to this session to see how XtremIO’s all-flash compare to traditional storage, explore use cases, see demos of XtremIO’s unique integrated copy data management (iCDM) stack integrate with vRealize Automation to accelerate SQL Server database application lifecycle management, and increase DBA productivities. See why traditional SQL Server best practices can be significantly simplified on XtremIO. We will focus on areas like VMFS/RDM, vSCSI adapters, queue depth, multi-pathing, number of LUNs, database file layout, tempdb placement, and more!


Automating Rapid and Space Efficient Provisioning of Oracle Database as a Service (DBaaS) using VMware vRealize Automation & EMC XtremIO for Mission [7569]

Automating Oracle Database as a Service (DBaaS) using VMware vRealize Automation Suite and EMC XtremIO provides accelerated deployment, elastic capacity, greater consolidation efficiency, higher availability, and lower overall operational cost and complexity for deploying Oracle Single Instance and Oracle RAC on VMware vSphere platform.


Automating Rapid and Space Efficient Provisioning of Oracle Database as a Service (DBaaS) using VMware vRealize Automation & EMC XtremIO [7572]

Automating Oracle Database as a Service (DBaaS) using VMware vRealize Automation Suite and EMC XtremIO provides accelerated deployment, elastic capacity, greater consolidation efficiency, higher availability, and lower overall operational cost and complexity for deploying Oracle Single Instance and Oracle RAC on VMware vSphere platform.


Achieving performance, scalability and efficiency for Oracle DBaaS infrastructure with vRealize Automation and EMC XtremIO all-flash scale-out storag [7626]

DBaaS has tremendous potential to transform how databases are provisioned, consumed, managed and decommissioned – simplifying day to day operations of DBAs, application owners, and developers. However, careful planning and agile infrastructure are the must for delivering DBaaS that performs consistently and scales predictably. Experts from VMware and EMC will discuss key considerations and best practices for successfully delivering Oracle DBaaS.



Achieving scalability, performance and efficiency for Oracle DBaaS infrastructure with vRealize Automation and EMC XtremIO all-flash storage [7599]

DBaaS has tremendous potential to transform how databases are provisioned, consumed, managed and decommissioned – simplifying day to day operations of DBAs, application owners, and developers. However, careful planning and agile infrastructure are the must for delivering DBaaS that performs consistently and scales predictably. Experts from VMware and EMC will discuss key considerations and best practices for successfully delivering Oracle DBaaS.



Tales from the Trenches: End-User Computing Architect Perspectives On Next Generation VDI [8222]

VDI architects from VMware and EMC XtremIO discuss the recent trends in desktop virtualization and role of all-flash storage in ushering in the new era of VDI. This technical session will examine Just-in-Time desktops and how to orchestrate desktop building blocks including Instant Clone Technology, AppVolumes and User Environment Management customizations to deliver personalized desktops. The role of XtremIO all-flash storage as the underlying high-speed engine will be discussed. Performance data for VDI with Windows 10 will be looked at in-depth with tips and recommendations for deployment and optimization.



Hardware as a Service: Case Study of automated provisioning of hardware into vSphere and Cloud Foundry environments [8609]

The Pivotal cloud building team called PEZ, has built an automated system where users submit a request in one end and fully configured vSphere with Pivotal Cloud Foundry on top comes out the other end. Full control is handed over from the bare metal hardware all the way up to the platform on top. All of this built using components of the EMC federation including vSphere, NSX (Network), PCF (PaaS) and XtremIO (Storage). This will be an exploration in automation showing how to inventory hardware, provision hardware via IPMI & PXE, install ESXi over PXE, install vCenter via OVFtool, configure vSphere components with PowerCLI, build networks with NSX over an API, provision storage over an API, and install PCF. All automated by robots. Technology covered will include IPMI, PXE, Foreman, ESXi 6.0, vCenter 6.0, NSX, vSAN, EMC ScaleIO, EMC XtremIO, and Pivotal Cloud Foundry. Tools mentioned include Python, PowerCLI, and lots and lots of magic. All materials will be published with details included. Portions of the project may be available as open source available on Github.



Cracking the Healthcare Code: The Avera Health Perspective [8828]

Healthcare IT organizations are facing a constant challenge of meeting provider needs to deliver quality patient care through always accessible systems. Healthcare providers require transformative solutions to innovate the desktop and application delivery using virtualization. Avera Health, a regional health system comprising more than 300 locations in 100 communities throughout US mid-west, faced several business challenges of aging PC’s in patient care areas, standardizing nurses’ stations, supporting BYO program with diverse devices, and remote access.  Partnering with VMware and EMC, Avera leveraged a combined solution of Horizon View, App Volumes, and User Environment Manager along with EMC XtremIO for a high performance desktop experience.  This strategy enabled Avera to create a standardized virtual desktop solution and consistent delivery of applications to providers allowing them to be prepared for a large scale deployment of Meditech to their hospitals and clinics. Providers are given easy access to applications and can enter data while preserving their need for mobility.  The result was extremely efficient workflows, improved patient safety, and reduced struggles with technology.   Attend this session to learn how VMware Horizon, App Volumes, User Environment Manager and EMC XtremIO can be implemented together to deliver a highly-available and scalable architecture worthy of the most demanding healthcare environments.


Under the covers of the virtualized data center [9011]

A modern businesses’ success is intrinsically linked to the operation of their End User Computing and Business Critical Application environments. The vast majority have already embraced the virtualized data center model and now enjoy the efficiencies realized through the decoupling of server hardware, operating system, and applications. The journey to further optimize your VMware hosted environment does not stop here though! In this session, product experts from Brocade and EMC will discuss what options are available to Solutions Architects and Data Center administrators to achieve further optimization and improvement in the delivery of virtualized IT services. This technical overview will focus on how to ensure you can achieve maximum usage of back-end hardware resources without compromising application performance. Verified vendor best practices to be used in the deployment of both virtual database and virtual desktop infrastructures will be presented and discussed. Opportunities for increased instrumentation across the entire I/O stack (storage and fabric) will be highlighted and examined – delivering performance you can trust with visibility into I/O performance from the application to virtual machine and through the physical layers. Vendor technologies to be discussed will include; Brocade (FC SAN, FabricVision, VM-ID, AMP), EMC (XtremIO All-Flash Storage), VMware (ESXi, Horizon, vROps

VMware Horizon 7 Instant Clones Technology on XtremIO

One of the new cool technologies that VMware have come up with recently is Instant Cloning aka “forking”.

Is was actually talked about way back in 2014 & 2015 but actually saw the light in Horizon 7 as part of just in time technology for VDI VMs.

You can read about it here


as someone who works a lot with VDI and helping our XtremIO customers to leverage the array in the best way they can, I wanted to examine the workload of instant clones VMs on a single XtremIO X-Brick, for this I provisioned 2500 VDI VMs running windows 10 (which in itself is very new to VDI as I only starting to see customers who are starting deploying it now) and office 2016, I didn’t take the easy route as both windows 10 + office 2016 add a significant load on the infrastructure (compute + storage) but since I want to be able to help with future deployments on XtremIO, I chose these two.

In order to generate the workload on the 2500 VMs, I used LoginVSI which is the industry standard to simulate VDI workloads.

The results can see below J

VPLEX GeoSynchrony 5.5 SP2 Is Out


I’m so happy this release is finally out, during the 3 years since XtremIO went GA (time passed by so fast..) we got many customers who wanted to leverage XtremIO with VPLEX, there are many reasons to do so but i would say that at least my favorite one is an Active/ Active virtualized data center based on VPLEX, as such, customers wanted to have the full feature set of XtremIO such as UNMAP and the ease of management, aka VIAS, happy to report these (and more) are now in!


Thin provisioning awareness for mirrors
– Enables host-based storage reclamation for local and distributed mirrors using UNMAP
–Enables UNMAP during Thin-> Thin migration
– Restricted to XtremIO and VMware ESXi (same as SP1)


. Reclaiming storage from the host using UNMAP to mirrored volumes
. Creating thin-provisioned mirrored virtual volumes
. Migrating thin-provisioned storage
. Noticing soft threshold crossing
. Handling out-of-space condition on thin mirrors

Device must be thin-capable
– 1:1 device geometry
– local and distributed RAID-1 mirrors are now supported
– underlying storage volumes must all come from XtremIO
• Virtual volume must still be thin-enabled
NEW: The virtual-volume provision –thin command can now be
used to create thin-enabled virtual volumes:
– provisions XtremIO storage volumes
– claims these storage volumes in VPLEX
– creates extents and devices atop the storage volumes
– creates virtual volumes and makes them thin-enabled


VPLEX uses a temporary mirror during migration
thin-enabled setting from source volume is
transferred to the target volume when the migration
is committed
• UNMAP is supported during and after migration when:
– the source and target are both thin-capable (XtremIO), and
– the source is thin-enabled


• If an XtremIO array crosses its configured resource
utilization soft threshold, it raises a Unit Attention:
• In SP2, VPLEX notices this Unit Attention and emits a call-home (limited to every 24 hours):
– scsi/170 WARNING: Thin Provisioning Soft Threshold
reached – vol <volName>
• This Unit Attention is not propagated to hosts on the
VPLEX front-end

When an out-of-space error is seen on the last healthy mirror
leg of a RAID1, VPLEX propagates the failure to the host
• When an out-of-space error is seen on a mirror leg while there
are other healthy mirror legs, VPLEX:
– Marks the affected mirror leg dead
– Prevents the affected mirror leg from being automatically
– Decreases RAID-1 redundancy, but host is unaffected
• After resolving resource exhaustion on XtremIO, admin must
manually resurrect the affected devices in VPLEX


• A new convenience command can be used to trigger
resurrection of dead storage volumes:
device resurrect-dead-storage-volumes –devices []


The virtual volume listing contains a thin-enabled
property that was a true/false boolean in SP1
• In SP2, this property has three possible states:
– enabled
– disabled (thin-capable but not enabled)
– unavailable (not thin-capable)


Two new statistics:
– unmap-ops
– unmap-avg-lat
• Each applicable to the following targets:
– host-init (I)
– fe-prt (T)
– fe-lu (L)
– fe-director (in the default monitors)


In SP1, thin-capable storage volumes were
automatically configured to use thin-rebuild
– i.e. skip the write to a rebuild target if the source and
target both hold zero data
• In SP2, this behavior has been reverted
– for XtremIO, normal rebuild gives better performance


In SP1, UNMAP was converted into a WRITE SAME 16
(identical semantics for XtremIO)
• In SP2 the VPLEX I/O is tagged internally as an
UNMAP, and we initiate an UNMAP to the XtremIO


VIAS now supports registering an XMS as an AMP managing multiple XtremIO arrays.
– No longer have to create one REST AMP per array as in VPLEX 5.5 and 5.5.1.
• During the REST AMP registration, it is no longer required to select the array. VIAS will automatically discover all the arrays.
• Under the cover, VIAS will either use the XtremIO REST API v1 or v2 based on the XMS version for provisioning.

Select the AMP to see the Managed Arrays on the right side.

Once the new REST AMP has been created,
managing virtual volumes with VIAS stays the same.
– There is no impact anywhere else in either the VPLEX GUI or CLI.

GUI Support for Thin Virtual Volume

  1. Overview

    A new attribute “Thin Enabled” was added to the virtual volume type.
    It can be seen in the ‘Virtual Volumes’ view and the property dialog.
    – The value can be unavailable, enabled or disabled.
    – Currently only virtual volumes created from XtremIO storage
    volumes can currently be Thin Enabled.
    – A new attribute “Thin Capable” (Yes/No for value) was added to the
    following object types:
    – Storage Volume
    – Device
    – Extent

    Changes in the GUI

    Storage Volumes Views

2. Property dialogs


A “Create thin virtual volumes” options was added to the “Create
Virtual Volumes” dialog. If the virtual volumes can’t be created as thin,
the operation will succeed but it will be thick instead.

4. Wizards

In the Provision from Pools/Storage Volumes, if all selected arrays are XtremIO, the user will have the option to create the virtual volumes as thin enabled with the addition of this new page in the wizard

Below you can see a recorded demo of the new UNMAP functionality

EMC AppSync 2.2 SP3 Is Out, Another Must Upgrade For XtremIO Customers


We have just GA’d a new service pack for EMC AppSync, as you probably already know, Copy Data Managemnet (CDM) is a key components of the XtremIO architecture, it is so big and so different than anything else that is out there that many customers are using XtremIO just for that.

But, what good does a CDM architecture and features if you don’t have an integral software the links between the storage array technology to the applications it needs to protect & repurpose. This is where Appsync comes in, if you are new to Appsync, I encourage you to start reading about it here first



EMC AppSync offers a simple, SLA-driven, self-service approach for protecting, restoring, and cloning critical Microsoft and Oracle applications and VMware environments. After
defining service plans (such as Gold, Silver, and Bronze), application owners can protect, restore, and clone production data quickly with item-level granularity by using the
underlying EMC replication technologies. AppSync also provides an application protection monitoring service that generates alerts when the SLAs are not met.
AppSync supports the following applications and storage arrays:
Applications — Oracle, Microsoft SQL Server, Microsoft Exchange, and VMware VMFS and NFS datastores, file systems, and Oracle application for NFS.
Storage — VMAX, VMAX 3, VNX (Block and File), VNXe, XtremIO, and ViPR Controller
Replication Technologies—VNX Advanced Snapshots, VNXe Unified Snapshot, SRDF, TimeFinder, SnapVX, RecoverPoint, XtremIO Snapshot, and ViPR Snapshot

ok, here’s what’s new in 2.2 SP3

lService pack full install.
Until AppSync 2.2.2, service pack was always an upgrade install. However, AppSync 2.2.3 supports full install.
Unmount callout script for AIX file system.
CLI enhancements include:
Repurposing refresh
Mount option for mounting all file system copies that are protected together (include_all_copies=true)
Expire option to remove a copy which has multiple associated copies
Unmount option to specify latest or oldest mounted copy
XtremIO specific fixes.
If you are on XtremIO 4.0.2, it is recommended that you upgrade to AppSync 2.2.3 because it includes critical XtremIO specific fixes.
Improved support for Linux MPIO user friendly names.
Supports VMAX Hudson HYPERMAX OS: 5977.809.784 with SMI-S 8.2.
Supports VMAX Indus HYPERMAX OS: 5977.691.684 with SMI-S 8.1.2.
Supports VMAX All Flash Arrays – 450F and 850F Models.
Supports RecoverPoint with SRM flag set in RecoverPoint Consistency Groups.
Supports RecoverPoint 4.4.
Supports RedHat Enterprise Linux, Oracle Linux, and CentOS 7.0, 7.1, and 7.2.
Supports IBM AIX 7.2.
Supports VERITAS VxVM 7.0, VCS 7.0, and VxFS 6.2, 7.0

Fixed issues (in 2.2 SP3)

URM00105205: Fixed issues with SUDO user permissions that were leading to host deployment failure.

URM00104744: Resolves the Mail Recipient field becoming null after rediscovery of Exchange host.

URM00104571: Added support to show Phase pits and corresponding events for the time period as configured by the user.

URM00105382: Addressed an issue of Mount failure when multiple Initiators from different ESX servers into a same Initiator groups for XtremIO.

URM00105041: Fixes the issue during create copy of exchange differential copy in VMAX V3 storage arrays.

URM00104870: This fixes timeout issue while creating remote copies of VNX file.

URM00105329: Addressed an issue where 1st gen copy of RAC database cannot be mounted if redo logs in separate ASM disk group.

URM00105066: Addresses an issue of RP bookmark copy not getting deleted from AppSync, even when the bookmark gets deleted from RPA.

URM00105450: Fixes optimistic lock exception when expire the VNX snapshots.

URM00105360: Fixes the issue of unable to get options “data only” and “log only” of Exchange database restore wizard.

URM00104609: Provided a fix to avoid indefinite wait by AppSync server for a response from Array.

URM00104799: Fixes a problem of AppSync Host Plugin hotfix hang on non-English machines.

URM00105077: Fixed an issue with SUDO user that leads to append extra sudo before the command execution.

URM00104906: Resolved timeout issue during RP bookmark creation.

URM00105258: Rectified a problem of Virtual Machine being detected as physical host.

URM00105278: Added a fix that will remove extra blank lines from the command output of powermt display.

URM00105281: Fixed an issue of Oracle 12c agent installation prevents discovery of hosts.

URM00105342: Fixed the mapping issue in case of Oracle DB created with UDEV rules.

URM00105629: Fix provided to validate RecoverPoint bookmarks for a CG before restore operation proceeds.

URM00105464: Fixes a device cleanup issue after unmounting a copy on AIX machine.

URM00105759: Fixed timeout issue while running fsck at mount time.

URM00105546: Fixed the discovery issue during mount.

URM00105501: Fixed PowerPath cleanup during unmount and enhanced CLI to mount all filesystem copies that are protected together.

URM00105607: Special characters in XtremIO 3.0 for folder name is handled for expire.

URM00105798: Addressed the unexpected Error while trying to delete an RPA.

URM00105538: Addressed performance issue with Oracle ASM mount with recove


VSI 6.8 Is Here, Here’s what’s new


I’m very happy to accounce we have just released Virtual Storage Integrator (VSI) 6.8!

If you have been living under a rock for the past 6 years, this is our vCenter (web) plugin that you can use to manage all of the EMC arrays, in the context of XtremIO there are so many things you can do with it that I’m using in a daily basis to manage my vSphere lab, oh, and it’s free which is always good.

If you want an overview of what the plugin does, you can start by reading here:



so, what’s new in 6.8

Yep, the number 1 request was to support VSI in a vCenter linked mode configuration, that is now supported.

Multiple vCenters Support – Overall

Multiple vCenters Support – VSI 6.8 Scope

All features related to XtremIO

Multiple vCenters Support – Preconditions

vCenters are configured in linked mode.
ŸDeploy VSI plugin to every single vCenter in
linked mode group.

Quality Improvements

Viewing XtremIO-based Datastore/RDM Disk Properties

Take several minutes to retrieve XtremIO-based datastore/RDM disk properties.
Optimize the algorithm to quickly match the underlying registered XtremIO array for datastore/RDM disk
Apply batch API/multiple threads to retrieve volumes/snapshots from XtremIO array.

Space Reclamation gets stalled

space reclamation task seen as ongoing in the vCenter tasks.
this has now been resolved.

VMware SRM Pairing fails

Upon trying to pair the SRM servers, you get a “token” error
this has now been resolved.

EMC World 2016 – The XtremIO Sessions

Wow, I can’t believe another year has passed since the last EMC World, time is flying by folks and that is a fact!

Here, at XtremIO we are very busy trying to have a solid agenda for you our customers, partners and SEs so you can leverage the event to come and hear about different topics, talk to the engineers and see how it all comes together.

Below you can see all the XtremIO sessions we have published so far and their dates, please notes that the dates for the sessions can still be change, looking forward to see you ALL there!

Registration for the event can be done from this url :


my session is the one highlighted in yellow.

Topic Title

Level (select one)
1. Beginner
2. Intermeditae
3. Advanced)

Session Day #1

Session Day #2


Abstract (60 words)

2016 All-Flash State of the Union: What’s New with XtremIO and How Customers Are Using It


Monday 8:30 – 9:30

Wed. 1:30 – 2:30

60 min.

This session provides an update on the latest XtremIO capabilities and how it is transforming customers’ workloads for agility, tremendous TCO and business benefits. With customer examples via real time data, we will discuss the use cases and benefits for consolidating mixed workloads across database, analytics, business apps.

Best Practices for Running Virtualized Workloads & Containers in the All-Flash Data Center


Monday 4:30 – 5:30

Wed 1:30 – 2:30

60 min.

Great, you customer have just purchased a shiny new all flash array (AFA), now what? In this session we will learn the reasons for one of the quickest revolutions that the storage industry has seen in the last 20 years, and how XtremIO can enable breakthrough capabilities for your server virtualization and private cloud deployments. We will go trough specific use case issues and how to overcome them. With lots of real-world tips and tricks, you can consolidate your most demanding virtualization workloads successfully and gracefully.

Building the Ultimate Database as a Service with XtremIO


Tuesday 3:00 – 4:00

Wed Noon – 1:00

60 min.

Database as a Service creates exciting new possibilities for developers, testers, business analysts and application owners. However, a successful deployment requires careful planning around database standardization and service automation, and an infrastructure that can safely consolidate multiple production and non-production workloads. This session will provide step-by-step guidance on getting your DBaaS project deployed on EMC XtremIO, the ideal platform for massive database consolidation.

Business Continuity, Disaster Recovery, and Data Protection for the All-Flash Data Center


Monday 3:00 – 4:00

Thurs 1:00 – 2:00

60 min.

Everybody knows that Flash is fast, but do you know the best way to protect your workload on a media that is capable of servicing more than 1M IOPS with tight RPO and RTO? This session will provide details of all the business continuity and disaster recovery solutions available for XtremIO customers. It will also include customer examples and recommend the best approaches for different scenarios. This session will cover the integration of XtremIO with ProtectPoint/Data Domain, VPLEX, RecoverPoint, AppSync and more.

Broken Promises, Buyer Beware: Special considerations for evaluating AFAs


Tuesday Noon – 1:00

Thurs 11:30 – 12:30

60 min.

Capacity & Performance are key factors in any storage purchase decision. But special consideration must be paid to them during any evaluation. In this session we’ll share best practices to be followed during evaluating an AFA to ensure a hiccup free production deployment.

Deploment Best Practices for Consolidating SQL Server and Integrated Copy Data Management


Monday 4:30 – 5:30

Wed. 8:30 – 9:30

60 min.

Examine SQL Server behavior on XtremIO’s all-flash array vs. traditional storage, explore use cases, see demos of XtremIO’s unique integrated copy data management (iCDM) stack which increases DBA productivity, and accelerates SQL Server database application lifecycle management. See why traditional SQL Server best practices can be significantly simplified with the deployment of XtremIO. We will focus on areas like storage deployment, queue depth, multi-pathing, number of LUNs, database file layout, tempdb placement, and more!

Learn from real life experts on how to deploy Big Data Analytics On an All-Flash Array implementation


Monday Noon – 1:00

Wed. Noon – 1:00

60 min.

This session provides an overview on the different architectural options available for customers to streamline their big data applications on a the XtremIO All-Flash Array. Learn from customers on the tradeoff and benefits of the different options. Understand how the different architectures can help customers scale / protect and simplify the storage infrastructure required to support their analytics workload. The presentation will discuss practical implications along with proven recommendations that storage professionals can walk away with.

Desktop Virtualization Best Practices: A Customer Perspective


Monday 8:30 – 9:30

Tues. 8:30 – 9:30

60 min.

Health Network Laboratories (HNL), a leading diagnostic test lab, leveraged a best-of-breed technology approach to produce stellar results moving its nearly 1,000 pathologists and scientists to VDI. Architected for a 10-year growth plan, the new VDI solution uses Citrix XenDesktop for brokering, EMC XtremIO for high performing, no-compromise desktops, Dell Wyse thin clients for access from anywhere, Imprivata for one-click single sign-on, Unidesk for application layering, and active-active data center with EMC VPLEX for business continuity. In this session, HNL CIO and his team will share what they learned in their journey to desktop virtualization nirvana.

Introducing XtremIO Integrated Copy Data Management and How It will transform Your Infrastructure, Your Operations, and Your Business Process Agility


Monday 1:30 – 2:30

Wed. 3:00 – 4:00

60 min.

Everybody knows that all-flash arrays are ideal for running database workloads – this is because flash is fast. But a database application is much more than just the production instance. Considerations in storage planning need to be made for additional copies of the database or application for things development and test, analytics and reporting, operations, and local recovery. Historically all-flash arrays have been too expensive to house these additional copies. Now, with innovations built into the EMC XtremIO all-flash array, these copies can be space-efficiently consolidated, operating at production levels of high performance, and managed easily in an automated fashion. The result is simplicity, lower cost application development, and faster software development cycles bringing agility and true transformation into your IT. Learn all about it with real customer examples in this session.

Next-Generation OpenStack Private Clouds with VMware and EMC XtremIO All-Flash Array


Monday Noon – 1:00

Thurs 8:30 – 9:30

60 min.

OpenStack Cloud deployments have long been notorious for being too complex and unpredictable, often requiring long implementation times and ongoing optimization. In this session, experts from VMware and EMC will discuss how VMware Integrated OpenStack (VIO), coupled with EMC XtremIO all-flash scale-out storage can dramatically simplify OpenStack implementations while delivering enterprise-class reliability, consistent predictable workload performance, and easy linear scalability.

Oracle Integrated Copy Data Management: Realizing the Power of XtremIO Virtual Copy Technology and EMC AppSync


Tuesday Noon – 1:00

Thurs. 10:00 – 11:00

60 min.

On average enterprise Oracle Database users create between 8 and 10 functional copies of each production database—and no enterprise Oracle Database user has but a single production database. Copy Data Management is one of the most troubling technical requirements in the enterprise today and it is first and foremost on the minds of Oracle Database Administrators. XtremIO Virtual Copy Technology leads the industry in terms of copy data management performance, flexibility and ease of use. To extend these XtremIO attributes to the self-service user, customers choose EMC AppSync. This session will introduce EMC AppSync and XtremIO Virtual Copy Technology and provide a case study covering ease of use and consistent, predictable performance across source volumes and virtual copies alike.

Game Changing SAP Best Practices for HANA and Traditional SAP, Consolidation, Converged Infrastructure, and iCDM on XtremIO


Monday 1:30 – 2:30

Thurs. 1:00 – 2:00

60 min.

99% of the Fortune 100 run their business on SAP. Almost 90% of those SAP architectures are built on an EMC SAP Data platform. Consolidation, reduced complexity & performance are primary focal points for these businesses. As is reducing cycles spent on infrastructure management to empower more focus on business innovation via SAP. Callaway Golf and Mohawk Industries are perfect examples of this mew Game-changing All-Flash driven SAP mantra. Callaway will show how they accelerated SAP performance In Production with EMC XtremIO All Flash Storage, by as much as 110%. Mohawk will show one of the worlds most advanced virtualized vHANA architectures on VCE Converged Infrastructure. Both customer teams will share best practices & lessons learned on reducing SAP costs via XtremIO Virtual Copy (XVC snapshots) & deduplication; accelerating performance and how EMC empowers their NextGen SAP strategies.

Tales From The Trenches: End-User Computing Architect Perspectives on Next Generation VDI


Tuesday 3:00 – 4:00

Thurs. 10:00 – 11:00

60 min.

In this session, VDI architects from VMware and EMC XtremIO will present the recent trends in desktop virtualization and role of all-flash storage in ushering the new era of VDI. Topics covered will include: (1) Just-in-Time desktops: How to orchestrate disparate desktop building blocks and application delivery schemas to deliver personalized desktops and the role of all-flash storage. (2) Application layering with VMware App Volumes: At scale performance with EMC XtremIO. (3) VDI with Windows 10: Key differences in storage access profiles from earlier Windows desktop OSes, optimization recommendations, and implications for all-flash storage.

XtremIO 101: All-Flash Architecture, Benefits, and Use Cases for Mixed Workload Consolidation


Tuesday 8:30 – 9:30

Thurs. 11:30 – 12:30

60 min.

This session provides an overview to the EMC XtremIO all-flash scale-out array and its design objectives. The architecture will be discussed and compared to other flash arrays in the market with the goal of helping the audience understand the unique requirements of building an all-flash array, the proper methodology for testing all-flash arrays, and architectural differentiation among flash array features that affect endurance, performance, and consistency across the most demanding mixed workload consolidations.

XtremIO iCDM Best Practices: Customter Panel


Monday 4:30 – 5:30 Trying to resched to not conflict with PR Session

Wed. 8:30 – 9:30

60 min.

Everybody knows that all-flash arrays are fast, but XtremIO enables much more than speeda and feed. Considerations in storage planning need to be made for additional copies of the database or application for things development and test, analytics and reporting, operations, and local recovery. Historically all-flash arrays have been too expensive to house these multiple copies of your production data. Now, with innovations built into the EMC XtremIO all-flash array, these copies can be space-efficiently consolidated, operating at production levels of high performance, and managed easily in an automated fashion. The result is simplicity, lower cost application development, reduce the required resources and have faster software development cycles. Learn from XtremIO customers how they implemented iCDM and how it transformed their IT to gain business agility and competitive edge.

Providing a persistent data volume to EMC XtremIO using ClusterHQ Flocker, Docker And Marathon


Containers are huge, that’s not a secret to anyone in the IT industry, customers are testing the waters and looking for many ways to utilize containers technologies, it is also not a secret that the leading vendor in this technology is docker.

But docker itself isn’t perfect yet, while it’s as trendy as trendy can get, there are many ways around the docker runtime to provide cluster management etc’..

It all started with a customer request some weeks ago, their request was “can you show us how do you integrate with docker, Marathon (to provide containers orchestration) and ClusterHQ Flocker to provide a persistent data volume”..sounds easy right? J

Wait a second, isn’t containers technologies supposed to be completely lossless, designed to fail and do not need any persistent data that will survive a failure in case a container dies??

Well, that’s exactly where customers are asking things that aren’t always part of the master design of containers technologies and where there is a gap, there is a solution..

Enter ClusterHQ with their Flocker product, taken from their website

What is Flocker?

Flocker is an open-source container data volume manager for your Dockerized applications.

By providing tools for data migrations, Flocker gives ops teams the tools they need to run containerized stateful services like databases in production.

Unlike a Docker data volume which is tied to a single server, a Flocker data volume, called a dataset, is portable and can be used with any container in your cluster.

Flocker manages Docker containers and data volumes together. When you use Flocker to manage your stateful microservice, your volumes will follow your containers when they move between different hosts in your cluster.

Container Manager and Storage Integrations

Flocker is designed to work with the other tools you are using to build and run your distributed applications. Flocker can be used with popular container managers or orchestration tools like Docker, Kubernetes, Mesos.

For storage, Flocker supports block-based shared storage such as Amazon EBS, or OpenStack Cinder so you can choose the storage backend that is best for your application. Read more about choosing the best storage backend for your application. You can also use Flocker to take advantage of pre-defined storage profiles that are offered by many storage providers. Find out more about Flocker’s storage profile support.

How is related to EMC XtremIO you might wonder, well, as the #1 sold AFA in the market, we are starting to get many requests like these and so, together with ClusterHQ, there is now a support for EMC XtremIO to provide this functionality.

If you want to see a full demo of a MySQL app failing over, look no further

I also wanted to give an huge thank you to Dadee Birgher from my team who set it all up in no time.