Description

Provides an overview of how Resource Usage is reported to CloudControl:

All CloudControl Usage In Base 2 ("gibibyte") Methodology

All CloudControl usage reporting related to "Gigabyte" (GB) is actually based on Base 2 (binary) methodology, meaning that 1 GB = 1024^3 bytes ("gibibyte"). This is true for all metrics, including storage.

When provisioning and managing assets in CloudControl, the system uses the same methodology. So if you provision a "100 GB" local disk, the system provisions a local disk of 100 x 1024^3 bytes = 107374182400 bytes.

Overview

CloudControl tracks usage of all Cloud assets such as Cloud Servers, Images, Networks, VLANs, and Sub-Administrators for billing and chargeback purposes. A daily total of usage per data center location is available to the Primary Administrator and users with the Reports role via the Summary Usage Report which summarizes usage data per location. The Detailed Usage Report report provides asset-level usage information for most of these usage elements, though some metrics use separate reports for asset-level details. This article describes each usage element and provides details as to how the usage is calculated.

CloudControl calculates usage once per day based on the calendar day of the timezone of the Geographic Region. The system will follow local daylight savings time, so in Regions where such changes occur, a day may consist of 23 or 25 hours on the day of the time change. For details on these timezones, see Introduction to Geographic Regions.

Most usage is computed on an hourly basis based on the existence of a specific type of Cloud asset and attributes of the asset (such as the number of GB or the service "plan"). For such metrics, usage is calculated down to the second for each asset in a given location. The usage data for all assets in a given location is then aggregated and the aggregate total is rounded up to the next integer to generate the location daily total used for usage and chargeback purposes and presented in the Summary Usage Report.

There are a few elements such as Bandwidth that are computed based on the total usage for the entire calendar day, usually in GB. For such metrics, usage is calculated separately for each asset in a given location and the all usage for all assets in that location is aggregated and rounded up to the next integer to generate the location daily total used for usage and chargeback purposes.

Organization Metrics - Sub-Administrators

Each Sub-Administrator added to an organization by a Primary Administrator (see How to Create a Sub-Administrator using the CloudControl UI) will generate one quantity of usage per hour to the Sub-Administrators element for each hour the sub-administrator exists.

Network Asset Metrics - Network Domains, VLANs, Public IPs, Cloud Networks, etc.

All Network assets generate usage whenever they are deployed. The usage metrics vary between MCP 1.0 and MCP 2.0:

MCP 2.0 Network Metrics

In MCP 2.0, each Network Domain can generate usage on four different usage elements:

  • Essentials Network Domain Hours, Advanced Network Domain Hours, Enterprise Network Domain Hours - The system calculates usage for each Network Domain based on the number of hours it exists and the type of Network Domain and assigns to the appropriate element. For example, an Advanced Network Domain will generate one Advanced Network Domain Hour for each hour it is deployed.
  • VLAN Hours - The system calculates usage for each VLAN based on the number of hours it exists as assigns usage to this element. For example, a VLAN will generate one VLAN Hour for each hour it is deployed.
  • Additional Public IPs - For each Public IP block assigned to an MCP 2.0 Network, the system calculates usage based on the two public IPv4 addresses included in the block and the number of hours the block exists. Each MCP 2.0 Public IP block will generate a quantity of 2 Additional Public IPs per hour for each hour that the IP block is deployed.

MCP 1.0 Network Metrics

In MCP 1.0, each Cloud Network can generate usage on three different usage elements:

  • Networks - The system calculates usage for each Cloud Network based on the number of hours it exists and assigns usage to this element. Each Cloud Network will generate one quantity of Networks for each hour it is deployed.
  • Additional Public IPs - For each additional Public IP block assigned a Cloud Network, the system calculates usage based on the number of public IPv4 addresses and the number of hours it exists. For example, a block of 2 IP addresses will generate a quantity of 2 Additional Public IPs for each hour that IP block is deployed. Note the "default" IP block included with each Cloud Network does not generate usage. Also, note MCP 1.0 uses IP blocks of different sizes.

Network Asset Bandwidth Usage

Network assets may generate bandwidth usage if there is IP traffic coming from inside or outside the asset. This usage is tracked under the usage elements Incoming Bandwidth (GB) and Outgoing Bandwidth (GB), where Incoming Bandwidth represents traffic directed into the network asset and Outgoing represents traffic leaving the network asset. These metrics are calculated differently in MCP 1.0 and MCP 2.0 locations:

  • In MCP 2.0 locations:
    • Incoming Bandwidth (GB) and Outgoing Bandwidth (GB) includes:
      • IPv4 and IPv6 traffic to/from the Public Internet.
      • IPv4 and IPv6 traffic between Network Domains (whether in the same or different data center locations)
        • NOTE: If a CPNC connection interconnects the Network Domains, this traffic is NOT counted as it is considered CPNC traffic as described below
    • Incoming Bandwidth (GB) and Outgoing Bandwidth (GB) excludes:
      • IPv4 and IPv6 traffic between VLANs in the same Network Domain
      • IPv4 and IPv6 traffic to/from a CPNC connection
      • IPv6 traffic to Cloud Backup infrastructure
  • In MCP 1.0 locations:
    • Metrics measure all traffic entering/leaving the Cloud Network, regardless of destination. Therefore, traffic related to IP communication with other Cloud Networks, CPNC connections, or Cloud Backup infrastructure is included in the metric.

The system identifies these values by querying the data center infrastructure once per day and identifying total bytes passed during the calendar day. Unlike other metrics which are rounded to the next highest integer, the system rounds bandwidth metrics to the nearest GB integer, so network assets with small amounts of traffic will report as 0 GB when they fall below the 500 MB threshold. If the system is unable to query the data center infrastructure at the time of measurement (i.e. due to network maintenance, outage, or a network interruption between the Geographic Region application instance and the data center location), it will also report 0 GB usage on that day.

Note that the Detailed Usage report described at How to Create a Detailed Usage Report will identify the metrics on a per Network Domain / Cloud Network level but does not identify further details as to what Cloud Server and/or network asset was the source or destination of the traffic. The source data does not include this information.

Cloud Server Metrics - CPU and RAM

Cloud Servers generate usage for the CPU and RAM usage metrics only when the server is running. There is no usage for these elements when the server is in a stopped state. If a server is stopped locally within the Guest OS, CloudControl will identify the stoppage within 5 minutes and cease calculation of usage for these metrics.

  • CPU Usage Metrics
    • The system calculates CPU usage based on the number of vCPU assigned to a server during the time when a server is running and the server's CPU Speed. Two separate sets of calculations are made. For billing and chargeback purposes, only one set of metrics is used. 
      • CPU Hours and High Performance CPU Hours - Calculated based on the total CPU Count, the CPU speed and amount of time the server is running. For example, a Server with 1 High Performance CPU that is running for 10.5 hours will generate 10.5 High Performance CPU hours. A server with 16 Standard CPUs that is running for 10 hours will generate 160 CPU Hours.
      • 1-12 Standard CPU Hours, 13+ Standard CPU Hours, 1-12 High Performance CPU Hours, and 13+ High Performance CPU Hours - Calculated in the same manner except that usage for the first 12 vCPU are assigned to the "1-12" element and all remaining CPU are assigned to the 13+ element. Using the same examples as above, a Server with 1 High Performance CPU that is running for 10.5 hours will generate 10.5 1-12 High Performance CPU hours and no 13+ High Performance CPU Hours. A server with 16 Standard CPUs that is running for 10 hours will generate 120 1-12 Standard CPU Hours (12 CPU x 10 Hours) and 40 13+ Standard CPU Hours (remaining 4 CPU x 10 Hours).
  • RAM Usage Metrics
    • The system calculates RAM usage based on the total amount of RAM assigned to a server during the time when a server is running. Again, two separate sets of calculations are made but only one set is used for billing and chargeback purposes. 
      • RAM Hours are calculated based on the total GB of RAM and the amount of time the server is running. For example, a Server with 200 GB RAM that is running for 10 hours will generate 2000 RAM Hours.
      • 1-24 GB RAM Hours,  25-48 GB RAM Hours, 49-128 GB RAM Hours, 129-256 GB RAM Hours, and 257+ GB RAM Hours - Calculated in the same manner except that usage is assigned to each element based on the amount of RAM within that range. Using the same example above, a 200 GB RAM server running for 10 hours generates:
        • 1-24 GB RAM Hours is = First 24 GB RAM * 10 Hours = 240 
        • 25-48 GB RAM Hours is = Next 24 GB RAM * 10 Hours = 240
        • 49-128 GB RAM Hours = Next 80 GB RAM * 10 Hours = 800
        • 129-256 GB RAM Hours = 72 GB RAM (200 GB RAM - 128) * 10 Hours = 720
        • 257+ GB RAM Hours = 0

Cloud Server and Client Image Metrics - Storage

Cloud Servers generate usage for the Storage usage metrics at all times, regardless of whether the server is started or stopped.

  • Storage Hours, Provisioned IOPS Storage Hours, High Performance Storage HoursEconomy Storage Hours, SSD Storage Hours - The system calculates storage usage based on the aggregate amount of storage assigned to a server in GB that is associated with a given Disk Speed. A given server can have multiple disks using different disk speeds so it may generate multiple types of storage usage.  For example, a Cloud Server with a 10 GB of Standard Storage disk and a 30 GB High Performance Storage disk that exists for 10 hours will generate 100 Standard Storage Hours (10 GB x 10 hours) and 300 High Performance Storage Hours (30 GB x 10 hours).
  • Provisioned IOPS Hours - Disks using a "Variable IOPS" disk speed (currently supported only for the Provisioned IOPS disk speed) generates IOPS usage in addition to the storage usage. IOPS usage is based the aggregate total of all of the chosen IOPS for all of the disks using the Variable IOPS speed. For example, a Cloud Server with a 100 GB Provisioned IOPS disk that is assigned 400 IOPS and exists for 10 hours will generate both 1000 Provisioned IOPS Storage Hours (100 GB x 10 hours) and 4000 Provisioned IOPS Hours (400 IOPS x 10 hours). 

There is one exception to the above rules. A newly created Snapshot Preview server generates only Storage Hours usage based on the aggregate amount storage, regardless of disk speed. It does not generate usage to the actual disk speed (including IOPS Hours if the disk is a Provisioned IOPS disk) until it is migrated to a normal Cloud Server. For more details, see Introduction to Cloud Server Snapshots.

All Client Images (whether created via clone or import) also generate Storage Hours based on the aggregate amount of all disk storage associated with the Image (regardless of disk speed) for as long as the Client Image exists. In the case of Images, the "speed" of any disk is just metadata to defining what speed the Image will default to using when deploying a server; the image itself is always stored on Standard Storage so only Storage Hours usage is incurred. For example, a Client Image with a total of 10 GB of "default" Standard Storage disks and 30 GB of default High Performance Storage disks that exists for 10 hours will generate 400 Storage Hours (40 GB x 10 hours).

Cloud Server Metrics - OS Units and Software Units

Cloud Servers may also generate usage based on the Operating System or Software installed on the Cloud Server.

  • OS Units - As described in more detail in Introduction to OS Units, each Operating System is a member of an OS Units Group. Each OS Group is defined to either incur a specific number of OS Units based on the number of vCPU or not to generate OS Units at all. If a server is a member of an OS Units Group that is  defined to incur OS Units, the appropriate number of OS Units are generated per hour based on the vCPU configuration of the server. This usage is generated regardless of whether the server is in a running or stopped state. The Detailed Usage report does not provide asset-level information about this usage. For asset-level details, see How to Create an OS Units Usage Report.
  • Software Units - As described in Introduction to Cloud ("Priced") Software, some OS Images include software for which usage is incurred. Each Priced Software package is defined to either incur a specific rate of Software Units based either on the on the number of vCPU or simply "per server". Separate rates are applied depending on whether the server is in a started or stopped state. If a Cloud Server has Priced Software, the appropriate number of Software Units are generated per hour based on the running state and, where necessary, the vCPU configuration of the server. The Detailed Usage report does not provide asset-level information about this usage. For asset-level details, see How to Create a Software Usage Report.

Cloud Server Optional Service Metrics - Cloud Server Snapshots, Cloud Monitoring, DRS for Cloud, and Security Groups

Cloud Servers provisioned with optional services incur other service-specific usage metrics.

  • Cloud Server Snapshots - Servers provisioned with the Cloud Server Snapshot feature incur usage to either Essentials Snapshot Hours or Advanced Snapshot Hours based on the server's Snapshot Service Plan.  This usage begins being incurred immediately upon service enablement before any snapshots are taken or stored. In addition, if Snapshot Replication is enabled for a Server with the Advanced Snapshot plan, Usage will appear under Advanced Replicated Snapshot Hours beginning from the time Replication was enabled on the Server.  Usage charges are based on the aggregate size of all of the local disks attached to the Cloud Server, and on the Service Plan selected (Essentials or Advanced), as well as Snapshot Replication, if enabled. The charges apply to the total size of the disks, not the amount used by the Operating System, as the stored snapshot includes the exact state of all local disks (even if unformatted) and any Disks that are replicated to a secondary location. Likewise, Advanced Replicated Snapshot Hours are incurred beginning from the time Replication was enabled on the Server. Usage stops when the service is disabled. Usage will continue to be incurred until Replication is fully disabled. The “(Snapshot Plan) Snapshot Service Hours” and "Advanced Replicated Snapshot Hours usage charges are in addition to the normal Economy/Standard/High-Performance usage charges for local disks that are described in Introduction to Cloud Server Local Storage ("Disks") and Disk Speeds:
      • EXAMPLE 1: A Cloud Server with 30 GB of Standard local storage and 50 GB of High-Performance local storage will generate 30 Storage Hours per hour and 50 High-Performance Hours per hour. If this Server is enabled for Cloud Server Snapshots, it will continue to generate 30 Storage Hours and 50 High-Performance Hours per hour but will now additionally generate 80 (Essentials or Advanced) Snapshot Service Hours so long as the Snapshot service is enabled. If disks are added or removed from the Cloud Server while enabled, the (Essentials or Advanced) Snapshot Service Hours will be adjusted automatically.
      • EXAMPLE 2: A Cloud Server enabled for Advanced Snapshots with 30 GB of Standard local storage and 50 GB of High-Performance local storage will generate 30 Storage Hours per hour and 50 High-Performance Hours per hour totaling 80 Advanced Snapshot Service Hours. If this Server is also enabled for Replicated Snapshots, it will generate an additional 80 Advanced Replicated Snapshot Hours as long as Snapshot Replication is enabled on the Server.
    • In addition, note that a "Snapshot Preview" Cloud Server that is newly created from a Snapshot (Local or Replicated) as described in How to Create a "Snapshot Preview" Cloud Server from a Successful Historical Snapshot incurs all local disks as Standard Disk Speed rates ("Storage Hours") while it is in Snapshot Preview mode. Local disks incur normal disk speed usage only after the server is migrated to a "Normal" server as described in How to Migrate a "Snapshot Preview" Server to a Normal Cloud Server.

      From a reporting perspective, details of Snapshot Usage (and Replicated Snapshot usage) are contained in the Snapshot Usage report as described in How to Generate a Snapshot Usage Report

  • Cloud Monitoring - Servers provisioned with the Cloud Monitoring feature incur usage to Essentials Monitoring Hours and Advanced Monitoring Hours based on the server's Cloud Monitoring Plan. For each hour a Cloud Server is provisioned with a given Cloud Monitoring Plan, it will generate one quantity to the associated usage element. For example, if a Cloud Server is provisioned with Advanced Cloud Monitoring, it will generate 24 Advanced Monitoring Hours per day so long as the server remains provisioned with that plan.
  • DRS for Cloud - Use of the DRS for Cloud service generates usage across two different elements:
    • DRS Source Server Hours - Each Source Server in a DRS Consistency Group contributes one hour of usage to this element for each hour it is a member of the Consistency Group. This usage is attributed to the data center location of the Source Server.
      1. Example: User creates a Consistency Group at Noon with two Source Servers in NA9 paired with two Target Servers in NA12. This configuration will generate 24 DRS Source Server Hours in NA9 on the first day (12 hours x 2 Source Servers = 24 Source Server Hours) and 48 DRS Source Server Hours/day thereafter until the Consistency Group is deleted. 
      2. Asset-level details on how this usage is computed can be identified through the DRS Pairs Usage report. This report identifies each Source Server / Target Server pair in a Consistency Group and how usage was computed for the Source Servers in those groups for each day. For details, see How to Generate a DRS Pairs Usage Report.
    • DRS Storage Hours - A Consistency Group includes a specific amount of DRS Journal storage in both the Source and Target locations. This usage is attributed to the DRS Storage Hours element separately in each location in the same manner that other disk storage usage is computed - one hour for each GB of DRS Journal storage.
      1. Example: User creates a Consistency Group at Noon with two Source Servers in NA9 paired with two Target Servers in NA12 and allocates 10,000 GB of DRS Journal Storage to the Consistency Group. On the first day, this configuration will generate 12,000 DRS Storage Hours in NA9 (12 hours x 10,000 GB = 12,000 DRS Storage Hours) and another 12,000 DRS Storage Hours in NA12. Thereafter, both NA9 and NA12 will be attributed 24,000 DRS Storage Hours/day until the Consistency Group is deleted. 
      2. Asset-level details for this usage appears on the Detailed Usage report. Each location's DRS Journal will appear on a separate row in the report with a column identifying the amount of DRS Journal storage and the attributed DRS Storage Hours.
    • Security GroupsSecurity Groups usage appears on the Summary and Security Group Usage reports under the usage element Security Group Hours.
      Security Group Hours start accruing starting from the time that either:
      1. A Cloud Server is added to a Server Security Group, or
      2. The first Server NIC is added to a VLAN Security Group
        Security Group Hours will continue to accrue until either the Cloud Server is removed from a Security Group or all Server NICs on a given Cloud Server are removed from Security Groups. usage element.
        From a reporting perspective, details of Security Group Usage are contained in the Security Group Usage report as described in How to Create a Security Group Usage Report

Cloud Backups

Cloud Servers enabled with the Cloud Backup feature will generate usage per the rules of the Cloud Backup product. Usage for on-premise Cloud Backups is also reported through the CloudControl usage system. For details, see Introduction to Cloud Backup Metering and Billing.