|
|
|
Storage |
|
|
|
Supported Storage
Details
|
DAS, SAS, iSCSI, NAS, SMB, FC, FCoE, openFCoE
XenServer data stores are called Storage Repositories (SRs), they support IDE, SATA, SCSI (physical HBA as well as SW initiator) and SAS drives locally connected, and iSCSI, NFS, SMB, SAS and Fibre Channel remotely connected.
Background: The SR and VDI abstractions allow advanced storage features such as Thin Provisioning, VDI snapshots, and fast cloning to be exposed on storage targets that support them. For storage subsystems that do not inherently support advanced operations directly, a software stack is provided based on Microsofts Virtual Hard Disk (VHD)
specification which implements these features.
SRs are storage targets containing virtual disk images (VDIs). SR commands provide operations for creating, destroying, resizing, cloning, connecting and discovering the individual VDIs that they contain.
Reference XenServer 7.0 Administrators Guide: http://bit.ly/2d8uA7C
Also refer to the XenServer Hardware Compatibility List (HCL) for more details.
|
=AU65
|
Yes
Support for NAS: Network Attached Storage, FC: Fibre Channel, iSCSI, FCoE. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes (limited for SAS)
Dynamic multipathing support is available for Fibre Channel and iSCSI storage arrays (round robin is default balancing mode). XenServer also supports the LSI Multi-Path Proxy Driver (MPP) for the Redundant Disk Array Controller (RDAC) - by default this driver is disabled. Multipathing to SAS based SANs is not enabled, changes must typically be made to XenServer as SAS drives do not readily appear as emulated SCSI LUNs.
|
=AU66
|
Yes
Multipath included. In addition to that Oracle offers the Connect Storage Plug-in to simplify Storage Management via Oracle VM Manager or EM13c Cloud Control. More info on: http://docs.oracle.com/cd/E64076_01/ and http://www.oracle.com/us/technologies/virtualization/ovm3-storage-connect-459309.pdf
|
|
|
Shared File System
Details
|
Yes (SR)
XenServer uses the concept of Storage Repositories (disk containers/data stores). These SRs can be shared between hosts or dedicated to particular hosts. Shared storage is pooled between multiple hosts within a defined resource pool. All hosts in a single resource pool must have at least one shared SR in common. NAS, iSCSI (Software initiator and HBA are both supported), SAS, or FC are supported for shared storage.
|
=AU67
|
Yes
OCFS2 included. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes (iSCSI, FC, FCoE)
XenServer 7.0 adds Software-boot-from-iSCSI for Cisco UCS
Yes for XenServer 6.1 and later (XenServer 5.6 SP1 added support for boot from SAN with multi-pathing support for Fibre Channel and iSCSI HBAs)
Note: Rolling Pool Upgrade should not be used with Boot from SAN environments. For more information on upgrading boot from SAN environments see Appendix B of the XenServer 7.0 Installation Guide: http://bit.ly/2d8BkSV
|
=AU68
|
Yes
Yes for SAN Boot with FC. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
No
While there are several (unofficial) approaches documented, officially no flash drives are supported as boot media for XenServer 7.0.
|
=AU69
|
Yes
Technically is possible but not officialy supported by Oracle. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
Virtual Disk Format
Details
|
vhd, raw disk (LUN)
XenServer supports file based vhd (NAS, local), block device-based vhd (FC, SAS, iSCSI) using a Logical Volume Manager on the Storage Repository or a full LUN (raw disk)
|
=AU70
|
Raw Images (*.img files)
If a disk is created from OVM Manager it is created by default as a QCOW image and if it is imported it is converted from the original format to QCOW Images. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
2TB
For XenServer 7.0 the maximum virtual disk sizes are:
- NFS: 2TB minus 4GB
- LVM (block): 2TB minus 4 GB
Reference: http://bit.ly/2dtaur1
|
=AU71
|
10 TB (if Virtual Disk on OCFS2), Maximum supported on the Guest OS/Filesystem (if Raw Disks)
For NFS there is no limit specified since it depends on the Backend Local Filesystem of the NFS Server. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
Thin Disk Provisioning
Details
|
Yes (Limitations on block)
XenServer supports 3 different types of storage repositories (SR)
1) File based vhd, which is a local ext3 or remote NFS filesystems, which supports thin provisioning for vhd
2) Block device-based vhd format (SAN based on FC, iSCSI, SAS) , which has no support for thin provisioning of the virtual disk but supports thin provisioning for snapshots
3) LUN based raw format - a full LUN is mapped as virtual disk image (VDI) so to is only applicable if the storage array HW supports that functionality
|
=AU72
|
Yes
Oracle provides the ability to create Sparse Disks and Thin Cloning for efficient disks usage based on the actual usage in addition to Non-Sparse Disks for full space allocation. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
No
There is no NPIV support for XenServer
|
=AU73
|
No
|
|
|
|
Yes - Clone on boot, clone, PVS, MCS
XenServer 6.2 introduced Clone on Boot
This feature supports Machine Creation Services (MCS) which is shipped as part of XenDesktop. Clone on boot allows rapid deployment of hundreds of transient desktop images from a single source, with the images being automatically destroyed and their disk space freed on exit.
General cloning capabilities: When cloning VMs based off a single VHD template, each child VM forms a chain where new changes are written to the new VM, and old blocks are directly read from the parent template. When this is done with a file based vhd (NFS) then the clone is thin provisioned. Chains up to a depth of 30 are supported but beware performance implications.
Comment: Citrixs desktop virtualization solution (XenDesktop) provides two additional technologies that use image sharing approaches:
- Provisioning Services (PVS) provides a (network) streaming technology that allows images to be provisioned from a single shared-disk image. Details: http://bit.ly/2d0FrQp
- With Machine Creation Services (MCS) all desktops in a catalog will be created off a Master Image. When creating the catalog you select the Master and choose if you want to deploy a pooled or dedicated desktop catalog from this image.
Note that neither PVS (for virtual machines) or MCS are included in the base XenServer license.
|
No
vCloud Director also linked clone capabilities but is not supported with this edition (requires Enterprise Plus)
VMwares virtual image sharing technology (vComposer or linked clones) is supported with VMwares virtual desktop solution (Horizon View).
Both Horizon View and vCD (as part of the vCloud Suites) are fee-based Vendor Add-Ons.
|
Yes
This technology is named Thin Cloning in Oracle VM provided by OCFS2 filesystem. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
SW Storage Replication
Details
|
No
There is no integrated (software-based) storage replication capability available within XenServer
|
=AU75
|
No
|
|
|
|
IntelliCache
XenServer 6.5 introduced a read caching feature that uses host memory in the new 64-bit Dom0 to reduce IOPS on storage networks, improve LoginVSI scores with VMs booting up to 3x Faster. The read cache feature is available for XenDesktop & XenApp Platinum users who have an entitlement to this feature.
Within XenServer 7.0 LoginVSI scores of 585 have been attained (Knowledge Worker workload on Login VS 4.1)
IntelliCache is a XenServer feature that can (only!) be used in a XenDesktop deployment to cache temporary and non-persistent operating-system data on the local XenServer host. It is of particular benefit when many Virtual Machines (VMs) all share a common OS image. The load on the storage array is reduced and performance is enhanced. In addition, network traffic to and from shared storage is reduced as the local storage caches the master image from shared storage.
IntelliCache works by caching data from a VMs parent VDI in local storage on the VM host. This local cache is then populated as data is read from the parent VDI. When many VMs share a common parent VDI (for example by all being based on a particular master image), the data pulled into the cache by a read from one VM can be used by another VM. This means that further access to the master image on shared storage is not required.
Reference: http://bit.ly/2d8Bxpm
|
No
vSphere Flash Read Cache is not enabled with this edition!
vSphere 5.5 introduced the vSphere Flash Read Cache that enables the pooling of multiple Flash-based devices into a single consumable vSphere construct called vSphere Flash Resource.
vSphere hosts can use the vSphere Flash Resource as vSphere Flash Swap Cache, which replaces the Swap to SSD feature previously introduced with vSphere 5.0. It provides a write-through cache mode that enhances virtual machines performance without the modification of applications and OSs.
At its core Flash Cache enables the offload of READ I/O from the shared storage to local SSDs, reducing the overall I/O requirements on your shared storage.
Documented maxima with vSphere 5.5:
- Virtual flash resource per host: 1
- Maximum cache for each virtual disk: 400GB
- Cumulative cache configured per host (for all virtual disks): 2TB
- Virtual disk size: 16TB
- Virtual host swap cache size: 4TB
- Flash devices (disks) per virtual flash resource: 8
There is also the Storage Accelerator (CBRC) feature that can be enabled (only) with VMware Horizon View (VDI).
View Storage Accelerator is an in memory (ESXi Server Memory) cache, caching common image blocks when reading virtual desktop images. It is applicable to stateless (non-persistent) as well as stateful (persistent) desktops and is transparent to the guest virtual machine/desktop. It does not require any special storage array technology and provides additional performance benefits when used in conjunction with Storage Array technologies.
http://www.vmware.com/files/pdf/techpaper/vmware-view-storage-accelerator-host-caching-content-based-read-cache.pdf
|
No
|
|
|
|
No
There is no specific Storage Virtualization appliance capability other than the abstraction of storage resources through the hypervisor.
|
=AU77
|
No
|
|
|
Storage Integration (API)
Details
|
Integrated StorageLink (deprecated)
Integrated StorageLink was retired in XenServer 6.2.
Background: XenServer 6 introduced Integrated StorageLink Capabilities. It replaces the StorageLink Gateway technology used in previous editions and removes the requirement to run a VM with the StorageLink components. It provides access to use existing storage array-based features such as data replication, de-duplication, snapshot and cloning. Citrix StorageLink allows to integrate with existing storage systems, gives a common user interface across vendors and talks the language of the storage array i.e. exposes the native feature set in the storage array. StorageLink also provides a set of open APIs that link XenServer and Hyper-V environments to third party backup solutions and enterprise management frameworks. There is a limited HCL for StorageLink supporting arrays. Details: http://bit.ly/2dqKRUT
|
=AU78
|
Yes
Oracle offers the Storace Connect Plug-in (included in Oracle VM subscriptions) that simplify storage management like LUN creation, removal, resize and other features directly by the use of Oracle VM Manager. More info on: http://www.oracle.com/us/technologies/virtualization/ovm3-storage-connect-459309.pdf
|
|
|
|
Basic
Virtual disks on block-based SRs (e.g. FC, iSCSI) have an optional I/O priority Quality of Service (QoS) setting. This setting can be applied to existing virtual disks using the xe CLI.
Note: Bare in mind that QoS setting are applied to virtual disks accessing the LUN from the same host. QoS is not applied across hosts in the pool!
|
Limited (basic, no SIOC)
Storage I/O Control is not enabled with this license but the basic disk shares and limit functionality still allows you to manage basic disk access priority for vm on the same host (not across hosts - so a single host with lower priority vm could congest I/O paths)
|
No
|
|
|
|
Networking |
|
|
|
Advanced Network Switch
Details
|
Yes (Open vSwitch) - vSwitch Controller
The Open vSwitch is fully supported
The vSwitch brings visibility, security, and control to XenServer virtualized network environments. It consists of a virtualization-aware switch (the vSwitch) running on each XenServer and the vSwitch Controller, a centralized server that manages and coordinates the behavior of each individual vSwitch to provide the appearance of a single vSwitch.
The vSwitch Controller supports fine-grained security policies to control the flow of traffic sent to and from a VM and provides detailed visibility into the behavior and performance of all traffic sent in the virtual network environment. A vSwitch greatly simplifies IT administration within virtualized networking environments, as all VM configuration and statistics remain bound to the VM even if it migrates from one physical host in the resource pool to another.
Details in the XenServer 7.0 vSwitch Controller Users Guide: http://bit.ly/2d98foz
|
No
vNetwork Distributed Switch functionality is not enabled with this license
|
Yes
Support for Centralized vSwitch included. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes (incl. LACP - New)
XenServer 6.1 added the following functionality, maintained with XenServer 6.5:
- Link Aggregation Control Protocol (LACP) support: enables the use of industry-standard network bonding features to provide fault-tolerance and load balancing of network traffic.
- Source Load Balancing (SLB) improvements: allows up to 4 NICs to be used in an active-active bond. This improves total network throughput and increases fault tolerance in the event of hardware failures. The SLB balancing algorithm has been modified to reduce load on switches in large deployments.
Background:
XenServer provides support for active-active, active-passive, and LACP bonding modes. The number of NICs supported and the bonding mode supported varies according to network stack:
• LACP bonding is only available for the vSwitch, while active-active and active-passive are available for both the vSwitch and Linux bridge.
• When the vSwitch is the network stack, you can bond either two, three, or four NICs.
• When the Linux bridge is the network stack, you can only bond two NICs.
XenServer 6.1 provides three different types of bonds, all of which can be configured using either the CLI or XenCenter:
• Active/Active mode, with VM traffic balanced between the bonded NICs.
• Active/Passive mode, where only one NIC actively carries traffic.
• LACP Link Aggregation, in which active and stand-by NICs are negotiated between the switch and the server.
Reference: http://bit.ly/2d8uA7C
|
Yes (no LACP Support)
Note: no distributed switch included with this license - so no LACP support - only standard NIC teaming;
vSphere has integrated NIC teaming capabilities. To utilize NIC teaming, two or more network adapters must be uplinked to a virtual switch (standard or distributed).
The key advantages of NIC teaming are:
- Increased network capacity for the virtual switch hosting the team.
- Passive failover in the event one of the adapters in the team goes down
There are various NIC load balancing (e.g. based on originating port, source MAC or IP hash) and failover detection algorithms (link status, Beacon probing) in vSphere, for details refer to: http://bit.ly/16Sm1ll
The maximum number of supported (and teamed) adapters per host varies by vendors and speed (max 32 for 1Gb and 8 for 10Gb)
New in vSphere 5.5. is the enhancement of the LACP capabilities.
The enhanced link aggregation feature provides choice in hashing algorithms and also increases the limit on number of link aggregation groups (22 new hashing algorithm options, 64 LAGs per host and 64 LAGs per VMware vSphere VDS, new workflows to configure LACP across a large number of hosts via templates)
Added in 5.1 was the Link Aggregation Control Protocol (LACP) support - which is a standards-based method to control the bundling of several physical network links together to form a logical channel for increased bandwidth and redundancy purposes.
LACP enables a network device to negotiate an automatic bundling of links by sending LACP packets to the peer. As part of the vSphere 5.1 release, VMware now supports this standards-based link aggregation protocol.
|
Yes
Support for Active/Backup Bonding, Load Balancer Bonding, LACP Bonding. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes (limited for mgmt. traffic)
VLANs are supported with XenServer. To use VLANs with virtual machines use switch ports configured as 802.1Q VLAN trunk ports in combination with the XenServer VLAN feature to connect guest virtual network interfaces (VIFs) to specific VLANs (you can create new virtual networks with XenCenter and specify the VLLAN IDs). The XenServer management interfaces cannot be assigned to a XenServer VLAN via a trunk port - to place management traffic on a desired VLAN the switch ports need to be configured to perform 802.1Q VLAN tagging/untagging (native VLAN or as access mode ports). In this case the XenServer host is unaware of any VLAN configuration.
XenServer 6.1 removed a previous limitation which caused VM deployment delays when large numbers of VLANs were in use. This improvement enables administrators using XenServer 6.x and later to deploy hundreds of VLANs in a XenServer pool quickly.
|
=AU82
|
Yes
Support for VLAN Interfaces included. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
No
XenServer does not support PVLANs.
Please refer to the Citrix XenServer Design: Designing XenServer Network Configurations guide for details on network design and security considerations http://bit.ly/2ctPL11
|
No
Private VLANs (sub-VLANs) are only supported with the virtual distributed switch.
|
No
|
|
|
|
Yes (guests only)
XenServer 6.1 introduced formal support for IPv6 in XenServer guest VMs (maintained with 6.5). Customers already used it with e.g. 6.0 but the 6.1 release notes list this as new official feature: IPv6 Guest Support: enables the use of IPv6 addresses within guests allowing network administrators to plan for network growth.
Full support for IPv6 (i.e. assigning the host itself an IPv6 address) will be addressed in the future.
|
=AU84
|
Yes
Supported. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
SR-IOV
XenServer 6.0 provided improved SR-IOV support for once specific card, Intels Niantic NIC, this card is not officialy supported for XenServer7.0
Generally with SR-IOV VF, functions that require VM mobility like live migration, workload balancing, rolling pool upgrade, High Availability and Disaster Recovery are not possible.
|
VMDirectPath (No SR-IOV)
|
Yes
Supported. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes
You can set the Maximum Transmission Unit (MTU) for a XenServer network in the New Network wizard or for an existing network in its Properties window. The possible MTU value range is 1500 to 9216.
|
=AU86
|
Yes
Supported. More info on: http://docs.oracle.com/cd/E64076_01/
|
|
|
|
Yes (TSO)
TCP Segmentation Offload can be enabled, see http://bit.ly/13e9WLi
By default, Large Receive Offload (LRO) and Generic Receive Offload (GRO) are disabled on all physical network interfaces. Though unsupported, you can enable it manually http://bit.ly/2djwZiZ
|
=AU87
|
Yes
Supported. More info on: http://docs.oracle.com/cd/E50245_01/index.html
|
|
|
|
Yes (outgoing)
QoS of network transmissions can be done either at the vm level (basic) by setting a ks/sec limit for the virtual NIC or on the vSwitch-level (global policies). With the DVS you can select a rate limit (with units), and a burst size (with units). Traffic to all virtual NICs included in this policy level (e.g. you can create vm groups) is limited to the specified rate, with individual bursts limited to the specified number of packets. To prevent inheriting existing enforcement the QoS policy at the VM level should be disabled.
Background:
To limit the amount of outgoing data a VM can send per second, you can set an optional Quality of Service (QoS) value on VM virtual interfaces (VIFs). The setting lets you specify a maximum transmit rate for outgoing packets in kilobytes per second.
The QoS value limits the rate of transmission from the VM. As with many QoS approaches the QoS setting does not limit the amount of data the VM can receive. If such a limit is desired, Citrix recommends limiting the rate of incoming packets higher up in the network (for example, at the switch level).
Depending on networking stack configured in the pool, you can set the Quality of Service (QoS) value on VM virtual interfaces (VIFs) in one of two places-either a) on the vSwitch Controller or b) in XenServer (using theCLI or XenCenter).
|
Limited (no NetIOC)
VMware Network I/O control requires a virtual distributed switch (vDS) which is not enabled with this license. You can still use basic traffic shaping (hard using average/peak bandwidth and burst) on the port group level of the standard switch
|
No
|
|
|
Traffic Monitoring
Details
|
Yes (Port Mirroring)
The XenServer vSwitch has Traffic Mirroring capabilities. The Remote Switched Port Analyzer (RSPAN) policies support mirroring traffic sent or received on
a VIF to a VLAN in order to support traffic monitoring applications. Use the Port Configuration tab in the vSwitch Controller UI to configure policies that apply to the VIF ports.
|
No
Port Mirroring is only available with the virtual distributed switch
|
Yes
Supported. More info on: http://docs.oracle.com/cd/E64076_01/ and http://docs.oracle.com/cd/E73210_01/EMLCM/GUID-2256CB37-5A22-4DDE-A3E3-9C87EC08FD81.htm#EMLCM93821
|
