VMware vCenter Chargeback Manager 2.0 released

VMware vCenter ChargeBack was released on Nov 30, 2011. It has quite few awaited enhancement and a bit of interface polishing. For those of you who did not hear of VMware vCenter Chargeback Manager or does not know why its required, I will brief you a bit before going into the details of the new releases. If you have been in the virtualization field for a while, then for sure you have heard of the VMs Sprawl(A large amount of virtual machines on your network without the proper IT management or control)  or Zombie VMs(VMs that are using resources but their role is no longer required).  For more info about Zombie VMs, you might want to check out my earlier post at: Zombie VMs.

Zombie VMs & Sprawl VMs have something in common, both of them waste your Virtual  Infrastructure resources extensively that could waste the saving you have done by moving from physical to virtual or even impact the performance you get from your Virtual Infrastructure. The larger the environment and the more VMs you spawn, Zombie VMs & Sprawl VMs get to be harder to control & increasingly important to address. Further when moving to Cloud Computing Sprawl & Zombie VMs seems to be easily getting out of control as the users get to spawn their VMs with easy and more agile fashion.

While products like VMware vCenter CapacityIQ & vKernel vOPS Capacity Manager can help you identify Sprawl & Zombie VMs in the environment, they don’t discourage users from spawning more of them as there is no cost associated with doing so. As every one know free drinks always run out much faster than paid ones. vCenter Chargeback Manager will be the product to discourage users/departments from spawning/over-sizing non required VMs by being able to show the costs to the Virtual Infrastructure to run the load of each department which can be used in a showback fashion to show the business which departments are wasting the most resources of the infrastructure which tend to discourage people from wasting resources as they are now held responsible for what they spawn. Further, VMware vCenter Chargeback Manager can be used in a Charge back fashion, where departments are charged for their usage which in turn cut on the free drinks and every one is held responsible for his spending and this usually a great way of reducing both VM Sprawl & Zombies to minimum.

The feedback of vCenter Chargeback Manager was quite good, but with one shared complains that I am not sure if they were a minus or pluses. Most complains have came about the bit of complexity to fully configure the product due to the amount of options it has. To a certain extent this can be an advantage, as you almost can not thing of building a chargeback or showback option scenario that you can not setup in there. Though in order to do that, tons of options had to be provided to the users to configure and adjust and use as required. VMware has listen to the customers feedback & in vCenter Chargeback Manager 2.0 has introduced an easier wizard to carry out some basic charge back models if that all the customer required, but in most cases customers have a complex pricing requirement and in order for the product to be able to deliver to the extensive requirements of different organizations it had to have all these options in place. I believe a great way to avoid disappointment is to get VMware PSO involved when you are testing or implementing vCenter ChargeBack Manager as they can implement it up to your requirements and get you up and running with easy in no time.

Further, the name of the product seems to confuse many customers. Many customers are just looking for a Showback product and does not require a chargeback. Well, don’t leave the product name fool you as it can deliver as extensive showback as it does for chargeback. I believe being named a chargeback a lot of customers are overlooking it when looking for a showback only solution which it can be a perfect fit for.

A good place to start when thinking about VMware vCenter Chargeback Manager 2.0 is to look at the sample reports below, & figure out how much of value could having such info add to your Virtual Infrastructure of Cloud.

Allocation Model: Prod_Org_allocation_Report_620

Pay As You Go Model: Dev_Org_payasyougo_Report_623[1]

Ok, now you have got a bit of info about VMware vCenter Chargeback Manager 2.0, you can find below a bit of info of what is new in this version of Chargeback as per the release notes:

• Automatic Report Scheduler
  In vCenter Chargeback Manager, you can define automatic report schedulers. These schedulers create report schedules for hierarchies and entities that match the criteria specified in the automatic report scheduler. The automatic report scheduler scans all the hierarchies and creates report schedules for the hierarchies and entities that match the specified criteria.

• Charge thin and thick provisioned virtual machines differently
  In this release, you can charge thin provisioned disks and thick provisioned disks differently. By default, thin provisioned disks are charged for the actual usage of the disk. You can override this behavior to charge the thin provisioned disks as thick provisioned in your billing policy.

• Cost variance and cost optimization
  vCenter Chargeback Manager provides a cost variance graph in the report dashboard that lets you analyze the day-to-day change in cost for a selected hierarchy or entity. The graph shows cost data for the last 30 days and that projected for the next 3 months in sets of 30 days each. vCenter Chargeback Manager lists different cost optimization opportunities, such as oversized virtual machines, undersized virtual machines, idle virtual machines, and powered off virtual machines. Cost variance and cost optimization is displayed only for the vCenter Servers that are integrated with VMware vCenter Operations.

• Showback Report
  A showback report lets you analyze how the cost is distributed among the entities based on a specified distribution policy. It is a configurable report that does not include any costs when it is generated. You can specify the total cost, fixed cost, and resource weight in the generated report to obtain the cost for each entity and for each resource per entity. The cost per entity is calculated based on the distribution policy that you select when generating the show back report.

• Apply fixed cost based on virtual machine state
  In vCenter Chargeback Manager, you can now define fixed costs that will be applied on an entity only for the duration for which the virtual machine in the entity is powered on.

• Tier-based storage costing
  vCenter Chargeback Manager lets you define storage tiers and configure cost on the tiers. All the datastores under a tier will be charged uniformly as per the cost configuration settings on the tier. You can also include VM storage profiles in a tier. The storage profiles defined in vSphere are synchronized and the datastores are automatically grouped according to their storage profiles. The cost configuration defined on a profile is applied on all the datastores that match the storage profile. Similarly, the cost configuration defined on a tier is applied to all the datastores or storage profiles added to the tier.

• Support for raw device mapping
  vCenter Chargeback Manager accounts for usage of hard disks that use raw device mapping. The corresponding cost and usage data is reported for the virtual machines that have disks using raw device mapping.

• Complete support for vSphere 5.0 and vCloud Director 1.5
  vCenter Chargeback Manager 2.0 supports the new features introduced in vSphere 5.0, such as VM storage profiles, and that introduced in vCloud Director 1.5, such as support for an SQL Server database.

• Partial support for IPv6
  In this release, vCenter Chargeback Manager supports IPv6 over IPv4 on an experimental basis. You can provide URLs with IPv6 IP addresses when connecting to vCenter Servers, LDAP Servers, vCenter Chargeback Manager databases, and vCenter Server databases. Ensure that the IPv6 IP address is enclosed in square brackets [ ] in the URL as per the standard convention.

• VM Instance Cost support for all hierarchies in vCenter Chargeback Manager
  In vCenter Chargeback Manager 2.0, you can define fixed cost pricing matrices for virtual machines based on vCPU count and memory bundles. Unlike earlier releases, this functionality is now available for all the hierarchies created in vCenter Chargeback Manager.

• Support for burstable billing or 95th percentile billing for the external network traffic in vCloud Director
  You can now calculate the cost for external network traffic in your vCloud Director setup based on the 95th percentile transfer rate. Starting with this release, vCenter Chargeback Manager introduces the external network transmit rate and external network receive rate resources and the Burstable Utilization resource attribute for these resources. You can define a billing policy that has these resource-attribute pair in the expression to calculate the cost based on the 95th percentile transfer rate. vCenter Chargeback Manager calculate the 95th percentile value based on the daily samples. That is, vShield Manager Data Collector runs a job that accounts for the samples for the last 24 hours and computes the 95th percentile value based on these samples.

• Support for overage charging for org vDCs in the Allocation Pool model of vCloud DirectorIn this release, vCenter Chargeback Manager provides the VMware Cloud Director apply overage charge on Allocation Pool vDC property for the Cloud Director Data Collector. This property must be set to true (default value is false) to account for the resource usage over and above the guaranteed reservation in vCloud Director. This is applicable only for CPU and memory. Also, you must use the VMware Cloud Director Overage Allocation Pool Cost Model to account for resource overage. You must, however, modify the cost model to define the base rate and overage rate for the resources.

• New cost models and billing policies for vCloud Director
  This release of vCenter Chargeback Manager introduces two new cost models and billing policies that are made available when you install the Cloud Director Data Collector.
  • VMware Cloud Director Actual Usage Cost Model: This cost model uses the new VMware Cloud Director Billing Policy – Actual Usage. This billing policy lets you calculate cost based on actual resource usage for all resources except count of networks, enabled IPSec VPN tunnel count, and NAT, DHCP, and firewall services. For these resources the allocation values defined in vCloud Director is used.
  • VMware Cloud Director Overage Allocation Pool Cost Model: This cost model uses the new VMware Cloud Director Billing Policy – Overage Allocation Pool. Only for this beta release, the billing policy calculates the overage cost for CPU based on actual usage and that for memory based on allocation. For external network transmit and external network receive the actual usage values are used and for all other resources the allocation values are used.

For more info, you can look directly at the VMware vCenter Chargeback Manager 2.0  release notes at: http://www.vmware.com/support/vcbm/doc/vcbm_2_0_0_release_notes.html

I hope this help some organizations reduce their Zombie & Sprawl VMs and make it easy for the rest of us to kill these kind of VMs before being created.