Skip to main content
Hitachi Vantara Knowledge

Generating chargeback reports

You can generate chargeback reports from the HCP System Management Console. A chargeback report contains current and historical usage statistics for HCP tenants and their namespaces. For each chargeback report, you specify the start and end dates of the report period, which is the time period that’s covered by the report, and select the reporting interval, which determines whether HCP generates hourly, daily, or total statistics for the specified report period.

Chargeback reports are a good source of information for system analysis, enabling you to adjust storage and bandwidth allocations based on usage patterns. These reports can also serve as input to billing systems that need to determine charges for capacity and bandwidth usage at the tenant or namespace level.

Chargeback reports cover only HCP tenants and their namespaces. They do not include the default tenant or its namespace.

About chargeback reports

A chargeback report contains aggregated namespace statistics for each HCP tenant that’s defined on the HCP system. For any given tenant, the total number of reads shown for a specific reporting interval is the total number of read operations that occurred during that interval in all the namespaces owned by that tenant.

Chargeback reports also contain aggregated HCP-tenant statistics for the HCP system as a whole. For example, the total number of reads for an HCP system shown for a specific reporting interval is the total number of successful read operations that occurred during that interval in each namespace that’s owned by each HCP tenant that’s defined on the HCP system. System statistics shown in chargeback reports do not include usage data for the default tenant and namespace.

For any HCP tenant that has granted system-level administrative access to itself, chargeback reports also contain statistics for the individual namespaces owned by that tenant.

When generated from the System Management Console, chargeback reports are in CSV format. Each line in a report contains the values for one namespace, for one tenant, or for the entire HCP system during a specific reporting interval (that is, during a specific hour, during a specific day, or during the entire report period).

If a chargeback report includes the namespaces that are owned by a tenant, the lines for those namespaces are ordered alphabetically below the lines for that tenant. The lines for the HCP system are at the end of the report. Multiple lines for a tenant, a namespace, or the HCP system are ordered in ascending chronological order.

Generating a chargeback report

When generating a chargeback report, you need to specify the report period and the reporting interval.

To specify the report period, you select start and end dates. The report period starts at 12:00:00 a.m. on the start date (shown as 00:00:00) and ends at 11:59:59 p.m. on the end date.

The earliest start date you can specify is either 30 days earlier than the current date or the date when the HCP system was installed (if the system was installed less than 30 days ago). The latest end date you can specify is the current date.

You can use the same date as the start date and the end date for a report period. In this case, the report period starts at 12:00:00 a.m. on the specified date and ends at 11:59:59 p.m. on that date.

The reporting interval you select determines whether the report contains hourly, daily, or total statistics:

  • An hourly report contains one set of statistics for each hour in the report period for which statistics are available.

    If the report period starts on the day when the HCP system was installed, the first set of statistics in the report covers the time the system first came online through the end of the last minute in that hour. For example, if the system came online at 2:30:20 p.m. on July 25th and the report period starts on July 25th, the first set of statistics in the report covers 2:30:20 p.m. through 2:59:59 p.m. on July 25th.

    If the report period ends on the current day, the last set of statistics in the report covers the time from the start of the current hour to the time when you requested the report. For example, if you request an hourly report at 2:30:15 p.m., the last set of statistics in the report covers 2:00:00 p.m. through 2:30:15 p.m. on the current day.

  • A daily report contains one set of statistics for each day in the report period, where a day is defined as the period from 12:00:00 a.m. through 11:59:59 p.m. on any given date.

    If the report period starts on the day when the HCP system was installed, the first set of statistics in the report covers the time the system first came online through the end of the last minute on that day. For example, if the system came online at 2:30:20 p.m. on July 25th and the report period starts on July 25th, the first set of statistics in the report covers 2:30:20 p.m. through 11:59:59 p.m. on July 25th.

    If the report period ends on the current day, the last set of statistics in the report covers 12:00:00 a.m. on the current day to the time when you requested the report. For example, if you request a daily report at 2:30:15 p.m., the last set of statistics in the report covers 12:00:00 a.m. through 2:30:15 p.m. on the current day.

  • A total report contains one set of statistics for the entire report period.

    If the report period starts on the day when the HCP system was installed, the statistics in the report cover the time the system first came online either through 11:59:59 p.m. on the last day of the report period or through the time when you requested the report on the currrent day.

    For example, if the system came online at 2:30:20 p.m. on July 25th and, on August 7th, you request a total report with a start date of July 25th and an end date of August 7th, the statistics in the report cover 2:30:20 p.m. on July 25th through 2:50:25 p.m. on August 7th.

    Statistics collected for the current hour may not include some reads and writes that have already occurred during the hour. After the hour is over, however, the statistics collected for it are complete.

    TipYou can use the HCP management API to generate chargeback reports that cover time periods that are longer than one month and are in XML, JSON, or CSV format. You can use this feature to create applications that generate chargeback reports at regular intervals and feed those reports to a billing system.

Generating a chargeback report

Before you begin

To generate a chargeback report, you need the monitor or administrator role.

Procedure

  1. In the top-level menu of the System Management Console, select Monitoring Chargeback.

  2. On the Chargeback page select the reporting interval for the type of storage usage report that you want to generate, and specify the start and end dates for the report period, in the Reporting Interval field, select Hour, Day, or Total to generate an hourly, daily, or total storage usage report, respectively.

  3. Use the Start Date field to specify the start date of the report period.

    • Enter the start date in the text field in mm/dd/yyyy format.
    • Click the calendar icon to display a monthly calendar. Then use the arrows at the top of the calendar to navigate to the month containing the start date that you want to define for the report period, and click that date to select it.
  4. Use the End Date field to specify the end date of the report period.

    • Enter the end date in the text field in mm/dd/yyyy format.
    • Click the calendar icon to display a monthly calendar. Then use the arrows at the top of the calendar to navigate to the month containing the end date that you want to define for the report period, and click that date to select it.
  5. To generate and download a complete, detailed chargeback report, click Download Report.

    By default, the name of the downloaded report file is Hourly-Chargeback-Report.csv, Daily-Chargeback-Report.csv, or Monthly-Chargeback-Report.csv, as applicable.

Chargeback statistics collection

Chargeback statistics either reflect a specific point in time or are dynamic. Point-in-time statistics are measurements taken at the end of a reporting interval, such as the used storage capacity for a namespace at the end of an hour. Dynamic statistics are measurements, such as the number of reads or writes to a namespace, that are accumulated over time.

HCP accumulates dynamic statistics on an hourly basis, starting at the beginning of each hour. So, for example, one statistic might represent the number of successful writes to a namespace that occurred between 11:00:00 and 11:59:59. Another might represent the number of successful writes to the same namespace that occurred between 12:00:00 and 12:59:59.

Chargeback report content

The first line of a chargeback report contains identifiers for the values in the subsequent lines. The following table describes each of these values and indicates whether it is point in time (P) or dynamic (D).

IdentifierTypeValue
bytesInD

The total number of bytes successfully written to the given namespace, to any of the namespaces owned by the given tenant, or to any of the HCP namespaces in the HCP system during the reporting interval.

The total number of bytes written includes all parts uploaded for multipart uploads regardless of whether the upload of the part has finished, the part was subsequently replaced, the multipart upload was aborted, or the part was not used in the completed multipart upload.

If data was compressed before being transmitted, this is the number of bytes before compression.

bytesOutD

The total number of bytes read from the given namespace, from any of the namespaces owned by the given tenant, or from any of the HCP namespaces in the HCP system during the reporting interval.

If data (including XML for directory listings) was compressed before being transmitted, this is the number of bytes before compression.

deletedN/A

One of:

  • true

    The namespace or tenant was deleted after the statistics in the set were collected.

  • false

    The namespace or tenant currently exists.

  • included

    For a tenant or the HCP system, the statistics in the set include values for one or more namespaces that were subsequently deleted.

deletesD

The total number of delete and purge operations successfully performed in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system during the reporting interval.

A delete of a multipart object counts as a single delete operation. The count of delete and purge operations does not include automatic deletions of replaced parts, parts of aborted multipart uploads, or unused parts of completed multipart uploads.

endTimeN/AThe end time of the reporting interval for the set of statistics in the line, in the same format as is used for the startTime value.
erasureCoded ObjectsPThe total number of erasure-coded objects and erasure-coded parts of multipart objects currently stored in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.
ingestedVolumePThe total size of the stored data and custom metadata, in bytes, before the data and custom metadata were added to the given namespace, to any of the namespaces owned by the given tenant, or to any of the HCP namespaces in the HCP system.

metadataOnly

Bytes

PThe total number of bytes of object data currently not stored for metadata-only objects in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.

metadataOnly

Objects

P

The total number of objects that are currently metadata-only in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.

A multipart object that is metadata-only counts as a single object.

multipartObject BytesPThe total number of bytes of object data in all the parts of multipart objects currently stored in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.
multipartObject PartsPThe total number of parts of multipart objects currently stored in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.
multipartObjectsPThe total number of multipart objects currently stored in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.
multipartUpload BytesP

The total number of bytes of object data in all the successfully uploaded parts of multipart uploads that are currently in progress in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.

This number does not include bytes of data in replaced parts of multipart uploads, parts uploaded for aborted multipart uploads, or unused parts of completed multipart uploads.

multipartUpload PartsP

The total number of successfully uploaded parts of multipart uploads that are currently in progress in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.

This number does not include replaced parts of multipart uploads, parts uploaded for aborted multipart uploads, or unused parts of completed multipart uploads.

multipartUploadsPThe total number of multipart uploads that are currently in progress in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.
namespaceNameN/A

The name of the namespace to which the set of statistics in the line applies.

In lines that contain tenant or systemwide statistics, this field has no value.

objectCountP

The number of objects in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system.

Each version of an object counts as a separate object. The object count does not include object versions that are delete markers. A delete marker is a special version of an object that indicates that a version of the object has been deleted. A delete marker has a version ID but does not have any data or metadata.

Each multipart object counts as a single object. Objects that are in the process of being created by multipart uploads are not included in the object count.

readsD

The total number of read operations performed in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system during the reporting interval.

A read of a multipart object counts as a single read operation.

startTimeN/A

The start time of the reporting interval for the set of statistics in the line, in this format:

yyyy-MM-ddhh:mm:ss

hh is hours on a 24-hour clock.

For example: 2017-10-07 14:32:47

storageCapacity

Used

P

The total number of bytes of primary running storage occupied by stored data for the given namespace, for all the namespaces owned by the given tenant, or for all the HCP namespaces in the HCP system. This includes object data, system and custom metadata, and any redundant data required to satisfy the applicable service plan.

Used storage capacity also includes storage occupied by the parts of in-progress multipart uploads that have already been written to HCP. Used storage capacity does not include replaced parts of multipart uploads, parts uploaded for aborted multipart uploads, or unused parts of completed multipart uploads.

systemNameN/A

One of:

  • The name of the domain associated with the data access network for the tenant that owns the namespace to which the set of statistics in the line applies
  • The name of the domain associated with the data access network for the tenant to which the set of aggregated statistics in the line applies
  • The name of the domain associated with the [hcp_system] network for the HCP system to which the set of aggregated statistics in the line applies
tenantNameN/A

Either:

  • The name of the tenant that owns the namespace to which the set of statistics in the line applies
  • The name of the tenant to which the set of statistics in the line applies

In lines that contain systemwide statistics, this field has no value.

tieredBytesP

Total number of bytes of object data currently stored on primary spindown storage (if the system includes spinddown storage), on S Series, or on extended storage for the given namespace, for all the namespaces owned by the given tenant, or for all the HCP namespaces in the HCP system.

On S Series only, the total number of bytes includes parts of in-progress multipart uploads regardless of whether the upload of the part has finished. This number does not include replaced parts of multipart uploads, parts uploaded for aborted multipart uploads, or unused parts of completed multipart uploads.

tieredObjectsP

The total number of objects with data currently stored on primary spindown storage (if the system includes spindown storage), on S Series, or on extended storage for the given namespace, for all the namespaces owned by the given tenant, or for all the HCP namespaces in the HCP system.

Each part of a multipart object counts as a separate object and, on S Series only, each part of an in-progress multipart upload counts as a separate object.

validN/A

The status of the set of statistics in the line. Possible values are:

  • true

    HCP successfully collected all statistics in the set.

  • false

    The statistics in the set do not reflect all the activity that occurred during the reporting interval. This may be due, for example, to one or more nodes being unavailable during that time, to a network failure, or to other hardware issues.

writesD

The total number of write operations successfully performed in the given namespace, in all the namespaces owned by the given tenant, or in all the HCP namespaces in the HCP system during the reporting interval.

Each upload of a part for a multipart upload counts as a separate write operation. This applies even if the part was subsequently replaced, the multipart upload was aborted, or the part was not used in the completed multipart upload.

Sample chargeback report

Here's an example of a daily chargeback report, where the report was requested for February 17, 2017, through February 18, 2017. The report shows statistics for the canada and europe tenants. Because system-level users have administrative access to the europe tenant, the report also shows statistics for the individual namespaces owned by the europe tenant. The report is shown as it would appear in a spreadsheet.

systemNametenantNamenamespaceNamestartTimeendTimeobjectCountIngestedVolumestorageCapacityUsedbytesInbytesOutreadswritesdeletestieredObjectstieredBytesmetadataOnlyObjectsmetadataOnlyByteserasureCodedObjectsmultipartObjectsmultipartObjectPartsmultipartObjectBytesmultipartUploadsmultipartUploadPartsmultipartUploadBytesdeletedvalid
hcp.example.comcanada 2/17/2017 0:00:002/17/2017 23:59:5915276947364276947364206315331247634615000000000000FALSETRUE
hcp.example.comcanada 2/18/2017 0:00:002/18/2017 23:59:59272977914243185724559147609364387227000000000000FALSETRUE
hcp.example.comeuropefinance2/17/2017 0:00:002/17/2017 23:59:5961342437211342709761239862638756111000000112793213889000FALSETRUE
hcp.example.comeuropefinance2/18/2017 0:00:002/18/2017 23:59:59151864967471865482245225302600120000023311132662065000FALSETRUE
hcp.example.comeuropehr2/17/2017 0:00:002/17/2017 23:59:597960936896215049609368007000000000000FALSETRUE
hcp.example.comeuropehr2/18/2017 0:00:002/18/2017 23:59:5915197073871973043210098019008000000000000FALSETRUE
hcp.example.comeurope 2/17/2017 0:00:002/17/2017 23:59:5913143853089143892480133595631001700000112793213889000FALSETRUE
hcp.example.comeurope 2/18/2017 0:00:002/18/2017 23:59:59302062041342062786566235104500200000023311132662065000FALSETRUE
hcp.example.com 2/17/2017 0:00:002/17/2017 23:59:592842080045342083984433991096233519573200000112793213889000FALSETRUE
hcp.example.com 2/18/2017 0:00:002/18/2017 23:59:5957503995558524851111153827138643872470000023311132662065000FALSETRUE

Chargeback page graphs

The Chargeback page contains four graphs. These graphs how information about the usage of the HCP system during the past 30 days (or since HCP was installed if that was less than 30 days ago).

The graphs are:

  • Writes (bytes)

    Shows the number of bytes of data written to the HCP system.

    The number of bytes written includes all parts written for multipart uploads regardless of whether the upload of the part has finished, the part was subsequently replaced, the multipart upload was aborted, or the part was not used in the completed multipart upload.

  • Writes (operations)

    Shows the number of successful write operations that targeted the HCP system.

    Each upload of a part for a multipart upload counts as a separate write operation. This applies even if the part was subsequently replaced, the multipart upload was aborted, or the part was not used in the completed multipart upload.

  • Reads (bytes)

    Shows the number of bytes of data read from the HCP system.

  • Reads (operations)

    Shows the number of read operations that targeted the HCP system.

    A read of a multipart object counts as a single read operation.

 

  • Was this article helpful?