VMware Server Virtualization FAQ

Rackspace offers two different solutions for DR:

• Replication Manager: Replication Manager (SRM) is an add-on service for VMware® Server Virtualization that utilizes VMware vCenter™ Site Recover Manager™ (SRM) to automate portions of the customer’s disaster recovery (DR) plan.

• VM Replication: VM Replication provides geographical redundancy and helps protect business-critical VMs in the event of a data center outage or unplanned downtime. Geographical redundancy is a key component of any sound DR strategy. VM Replication helps protect and recover VMware Server Virtualization VMs by easily and cost-effectively replicating VMs between our data centers.

With VMware Server Virtualization, you can use the MyRackspace Web Portal to easily administer your hosted environment and resources. Offload the maintenance and monitoring of the infrastructure, VMware stack, and Guest OS layer to your virtual IT team consisting of VMware Certified Professionals and experts in storage, networking, and infrastructure. This way you can focus more on your core business. With industry-leading SLAs that include 100% network uptime guarantees and one-hour hardware replacement guarantees, we can help make sure your mission-critical applications stay up and running.

No. However, you can view the performance and other statistics through the MyRackspace® portal. Contact your Rackspace VMware support team for more details.

No. Rackspace does not provide API access. If you are interested in gaining access to the software, speak to your Rackspace account team about Rackspace Private Cloud powered by VMware.

You can request this through an action in the MyRackspace Portal.

Yes, they are available for an additional service fee by using VM Recovery.

VMware Server Virtualization does not currently allow VMware vSphere® Storage vMotion® so you cannot enable fault tolerance.

You can suspend the utility billing of virtual machines when they are not in use by powering down VMs through the MyRackspace Portal. By powering off a VM from within the OS, you not only generate an alert ticket for support to investigate, but the VM continues billing at the daily utility rate. To power down VMs through the portal, see Powering a virtual machine on and off.

Rackspace can back up VMs through Rackspace Managed Backup service. Guest OS agent-based backups are available along with image-level backup through the VM Recovery add-on Service.

A snapshot is a point-in-time delta file to track all changes to a virtual machine. Snapshots give you the ability to roll back (Windows patching at the disk level), but they are not a permanent backup. You generally should not keep them for more than 72 hours.

• Ordering new or multiple VMs
• Change the resources of a VM
• Resize VMs
• Request re-imaging of a VM
• Cloning and copying VMs
• Power on or off and reset VMs
• Performance metrics for VMs, hypervisors, and clusters (CPU/RAM utilization, network, disk utilization)
• Requesting deletion of VMs

Less is better. It is more important to size VMs closer to what they need as opposed to loading in extra capacity that you might not use. Assigning too many CPUs slows performance down. Begin with two CPUs, and go to four if necessary. Very rarely start with four or more. Regardless of vendor recommendation for a physical environment, virtual CPUs do not map 1 to 1 to physical CPUs. It's a completely different architecture.

This overhead is necessary to perform functions like cloning, snapshotting, or vMotion.

To add RAM or CPU to your VM, request a ticket with your Rackspace VMware support team through the MyRackspace Portal.

• Clone: A clone is a one-to-one image copy of an existing VM. It acts independently from a parent VM, but it is an exact copy of the parent VM when it is initially created. You can create new VMs from the clones. You can create one clone per VM on the Rackspace infrastructure. The clone is also stored on the Rackspace infrastructure, so it doesn't use up your storage allocation. While clones can act and operate independently of the parent, snapshots cannot.

• Template: A template is a clone that is designed for the deployment of future VMs. A template is also stored on the Rackspace storage infrastructure. You can create one template per VM.

• Snapshots are stored with the parent disk of VM and consume your allocated storage infrastructure. You must carefully manage snapshots to ensure that they don’t consume all the disk space.
• You must have sufficient additional storage to store clones and templates.

You can expand your virtual disk, but you can't shrink it. To expand it, submit a ticket and specify the virtual disk that you want expanded.

You can use the MyRackspace Portal to view hypervisor and virtual machine performance. For instructions on how to view performance, see the Viewing Hypervisor and Virtual Machine performance section of the Server Virtualization Customer Handbook.

For information on how to create and delete clones or templates, see the Copy and cloning virtual machines section of the Server Virtualization Customer Handbook.

For information on how to create and delete snapshots, see the Snapshotting a virtual machine section of the Server Virtualization Customer Handbook.

Due to our licensing agreements with vendors, we cannot provide a licensed VM image. If you want to have a copy of a Rackspace VM, we can export the VM and remove the licensing. Speak to your VMware support team for more information.

Yes. The **Available RAM (or CPU)** field in the portal takes into account all VMs, whether powered down or not. However, powered-down VMs prevent the VM from consuming resources on the hypervisor.

You can have one clone, one template, and unlimited snapshots. Snapshots can be "unlimited" in that they take up to the original size of the original disk. The snapshots take up customer storage and do count toward the required 15% overhead.

IMPORTANT We do not recommend having more snapshots than is necessary. Snapshots can consume your resources and cause performance issues. Speak to your VMware support team for more information.

Deployment times can vary based on the request, availability of resources or IP addresses, and storage or compute resources (CPU and RAM) and hypervisors.