CDNI Capacity Capability Advertisement Extensions

While delegating traffic from an upstream CDN (uCDN) to a downstream CDN (dCDN), it is important to ensure that an appropriate amount of traffic is delegated. To achieve that, this specification defines a feedback mechanism to inform the delegator how much traffic may be delegated. The traffic level information provided by that interface will be consumed by services, such as a request router, to inform that service's traffic delegation decisions. The provided information is advisory and does not represent a guarantee, commitment, or reservation of capacity. This document defines and registers CDNI Payload Types (as defined at section 7.1 of ). These Payload types are used for Capability Objects added to those defined at section 4 of , which are required for the Open Caching Capacity Insights Interface.

The following terms are used throughout this document: CDN - Content Delivery Network Additionally, this document reuses the terminology defined in . Specifically, we use the following CDNI acronyms: uCDN, dCDN - Upstream CDN and Downstream CDN, respectively

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here.

To enable information exchange between a uCDN and a dCDN regarding acceptable levels of traffic delegation, the following process has been defined: In normal operation a uCDN will communicate with a dCDN, via an interface, to collect and understand any limits that a dCDN has set forth for traffic delegation from a uCDN. These limits will come in the form of metrics such as bits per second, requests per second, etc. These limits can be thought of as Not to Exceed (NTE) limits. The dCDN should provide access to a telemetry source of near real-time metrics that the uCDN can use to track current usage. The uCDN should compare its current usage to the limits the dCDN has put forth and adjust traffic delegation decisions accordingly to keep current usage under the specified limits. In summary, the dCDN will inform the uCDN of the amount of traffic that may be delegated. Additionally, it will provide a telemetry source aligned with this limit, allowing the uCDN to monitor its current usage against the advertised value. Having a limit and a corresponding telemetry source creates an unambiguous definition understood by both parties. Limits that are communicated from the dCDN to the uCDN should be considered valid based on the TTL (Time To Live) provided by a mechanism of the underlying transport, e.g., an HTTP Cache-Control header. The intention is that the limits would have a long-lived TTL and would represent a reasonable peak utilization limit that the uCDN should target. If the underlying transport does not provide a mechanism for the dCDN to communicate the TTL of the limits, the TTL should be communicated through an out-of-band mechanism agrred between the dCDN and uCDN.

Section 5 of describes the FCI Capability Advertisement Object, which contains a CDNI Capability Object as well as the capability object type (a CDNI Payload Type). The section also defines the Capability Objects per such type. Below, we define two additional Capability Objects. Note: In the following sections, the term "mandatory-to-specify" is used to convey which properties MUST be included when serializing a given capability object. When mandatory-to-specify is defined as a "Yes" for an individual property, it means that if the object containing that property is included in an FCI message, then the mandatory-to-specify property MUST be included.

The Telemetry Capability Object advertises a list of telemetry sources made available to the uCDN by the dCDN. In this document, telemetry data is being defined as near real-time aggregated metrics of dCDN utilization, such as bits per second egress, and is specific to the uCDN and dCDN traffic delegation relationship. Telemetry data is uniquely defined by a source ID, a metric name, and the footprints that are associated with an FCI.Capability advertisement. When defining a CapacityLimit, the meaning of a limit might be ambiguous if the uCDN and dCDN are observing telemetry via different data sources. A dCDN-provided telemetry source that both parties reference serves as a non-ambiguous metric for use when comparing current usage to a limit. Telemetry data is important for making informed traffic delegation decisions. Additionally, it is essential in providing visibility of traffic that has been delegated. In situations where there are multiple CDN delegations, a uCDN will need to aggregate the usage information from any dCDNs to which it delegated when asked to provide usage information, otherwise the traffic may seem unaccounted for. Example: A Content Provider delegates traffic directly to a uCDN, and that uCDN delegates that traffic to a dCDN. When the Content Provider polls the uCDN telemetry interface, any of the traffic the uCDN delegated to the dCDN would become invisible to the Content Provider unless the uCDN aggregates the dCDN telemetry with its own metrics. Property: sources Description: Telemetry sources made available to the uCDN. Type: A JSON array of Telemetry Source objects (see ). Mandatory-to-Specify: Yes.

The Telemetry Source Object is built of an associated type, a list of exposed metrics, and type-specific configuration data. Property: id Description: An identifier of a telemetry source. The ID string assigned to this Telemetry Source MUST be unique across all Telemetry Source objects in the advertisement containining this Telemetry Source Object. The ID string MUST remain consistent for the same source reference across advertisements. Type: String. Mandatory-to-Specify: Yes. Property: type Description: A valid telemetry source type. See . Type: String. Mandatory-to-Specify: Yes. Property: metrics Description: The metrics exposed by this source. Type: A JSON array of Telemetry Source Metric objects (see ). Mandatory-to-Specify: Yes. Property: configuration Description: a source-specific representation of the Telemetry Source configuration. For the generic source type, this configuration format is defined out-of-band. For other types, the configuration format will be specified in a yet to be defined telemetry interface specification. The goal of this element is to allow for forward compatibility with a formal telemetry interface. Type: A JSON object, the structure of which is specific to the Telemetry Source and outside the scope of this document. Mandatory-to-Specify: No.

At the time of this writing, the registry of valid Telemetry Source Object types is limited to a single type: Generic (see ). Source Type Description genericAn object which allows for advertisement of generic data sources

The Telemetry Source Metric Object describes the metric to be exposed. Property: name Description: An identifier for this metric. This name MUST be unique among metric objects within the containing Telemetry Source. The name MUST remain consistent for the same source reference across advertisements. Type: String. Mandatory-to-Specify: Yes. Property: time-granularity Description: The time, in seconds, representing the metric data. For example, a value representing the last 5 minutes would have a time-granularity of 300. Type: Unsigned Integer. Mandatory-to-Specify: No. Property: data-percentile Description: The percentile calculation the data represents, i.e., 50 percentile would equate to the median over the time-granularity. Lack of a data-percentile indicates that the data MUST be the mean over the time representation. Type: Unsigned Integer. Mandatory-to-Specify: No. Property: latency Description: Time in seconds that the data is behind real-time. This is important to specify to help the uCDN understand how long it might take to reflect traffic adjustments in the metrics. Type: Unsigned Integer. Mandatory-to-Specify: No.

The following shows an example of Telemetry Capability including two metrics for a source, that is scoped to a footprint.

] } ] } ]]>

The CapacityLimits Capability Object enables the dCDN to specify traffic delegation limits to a uCDN within an FCI.Capabilities advertisement. The limits specified by the dCDN will inform the uCDN on how much traffic may be delegated to the dCDN. The limits specified by the dCDN should be considered Not To Exceed (NTE) limits. The limits should be based on near real-time telemetry data that the dCDN provides to the uCDN. In other words, for each limit that is advertised, there should also exist a telemetry source which provides current utilization data against the particular advertised limit. Property: limits Description: A collection of CapacityLimit objects. Type: A JSON array of CapacityLimit objects (see ). Mandatory-to-Specify: Yes.

A CapacityLimit object is used to represent traffic limits for delegation from the uCDN towards the dCDN. The limit object is scoped to the footprint associated with the FCI capability advertisement encompassing this object. Limits MUST be considered using a logical "AND": a uCDN will need to ensure that all limits are considered rather than choosing only the most specific. Property: limit-type Description: The units of maximum-hard and maximum-soft. Type: String. One of the values listed in . Mandatory-to-Specify: Yes. Property: id Description: Specifies an identifier associated with a limit. This MAY be used as a relational identifier to a specific CapacityLimit Object. If specified, this identifier MUST be unique among specified identifiers associated with any other CapacityLimit objects in the advertisement containing this CapacityLimit Object. Type: String. Mandatory-to-Specify: No. Property: maximum-hard Description: The maximum unit of capacity that is available for use. Type: Unsigned Integer. Mandatory-to-Specify: Yes. Property: maximum-soft Description: A soft limit at which a uCDN SHOULD reduce traffic before hitting the hard limit. This value MUST be less than the value of maximum-hard. If this value is not specified, it is equal to the value of maximum-hard. Type: Unsigned Integer. Mandatory-to-Specify: No. Property: current Description: Specifies the current usage value of the limit. It is NOT RECOMMENDED to specify the current usage value inline with the FCI.CapacityLimits advertisements as it will reduce the ability to cache the response, but this mechanism exists for simple use cases where an external telemetry source cannot be feasibly implemented. The intended method for providing telemetry data is to reference a Telemetry Source object (see ) to poll for the current usage. Type: Unsigned Integer. Mandatory-to-Specify: No. Property: telemetry-source Description: Mapping of each particular limit to a specific metric with relevant real-time data provided by a telemetry source. Type: CapacityLimitTelemetrySource object (see ). Mandatory-to-Specify: No.

Below are listed the valid capacity limit-types registered in the CDNI Capacity Limit Types registry. The values specified here represent the types that were identified as being the most relevant metrics for the purposes of traffic delegation between CDNs. Limit Type Units egressBits per second requestsRequests per second storage-sizeTotal bytes storage-objectsCount sessionsCount cache-sizeTotal bytes

The CapacityLimitTelemetrySource Object refers to a specific metric within a Telemetry Source. Property: id Description: Reference to the "id" of a telemetry source defined by a Telemetry Capability object as defined in . Type: String. Mandatory-to-Specify: Yes. Property: metric Description: Reference to the "name" property of a metric defined within a telemetry source of a Telemetry Capability object. Type: String. Mandatory-to-Specify: Yes.

The following shows an example of an FCI.CapacityLimits object.

" ] } ] } ]]>

This document requests the registration of two additional payload types to the Content Delivery Network Interconnection (CDNI) Parameters "CDNI Payload Types" registry: Payload Type Specification FCI.TelemetryRFCthis FCI.CapacityLimitsRFCthis [RFC Editor: Please replace RFCthis with the published RFC number for this document.]

Purpose: The purpose of this Payload Type is to list the supported telemetry sources and the metrics made available by each source. Interface: FCI. Encoding: See .

Purpose: The purpose of this Payload Type is to define Capacity Limits based on utilization metrics corresponding to telemetry sources provided by the dCDN. Interface: FCI. Encoding: See .

IANA will add the following new registry to the "Content Delivery Network Interconnection (CDNI) Parameters" group at https://www.iana.org/assignments/cdni-parameters: Registry Name: CDNI Telemetry Source Types Registry Description: The CDNI Telemetry Source Types registry defines the valid values for the "type" property of the Telemetry Source object defined in . Registration Procedure: The registry follows the Specification Required policy as defined in . The Designated Expert should consider the following guidelines when evaluating registration requests: The new type definition does not duplicate existing types. The review should verify that the telemetry source is applicable to the CDNI use cases and that the description is clear and unambiguous. The registration is applicable for general use and not proprietary. The "configuration" property has a fully specified object definition with a description of each defined property. The following values will be registered: Source Type Specification genericRFCthis

Purpose: The purpose of this Telemetry Source Type is to provide a source-agnostic telemetry type that may be used for generic telemetry source advertisement. Usage: See .

IANA will add the following new registry to the "Content Delivery Network Interconnection (CDNI) Parameters" group at https://www.iana.org/assignments/cdni-parameters: Registry Name: CDNI Capacity Limit Types Registry Description: The CDNI Capacity Limit Types registry defines the valid values of the "limit-type" property of a CapacityLimit object defined in . Registration Procedure: The registry follows the Specification Required policy as defined in . The Designated Expert should consider the following guidelines when evaluating registration requests: The new capacity limit type does not duplicate existing entries. The submission has a defined purpose. The newly defined capacity limit type should be clearly justified in the context of one or more CDNI use cases. The description of the capacity limit type is well-documented and unambiguous. The following values will be registered: Capacity Limit Type Units Specification egressBits per secondRFCthis requestsRequests per secondRFCthis storage-sizeTotal bytesRFCthis storage-objectsCountRFCthis sessionsCountRFCthis cache-sizeTotal bytesRFCthis Usage: See .

This specification is in accordance with the CDNI Request Routing: Footprint and Capabilities Semantics. As such, it is subject to the security and privacy considerations as defined in Section 7 of .

The authors would like to express their gratitude to the members of the Streaming Video Technology Alliance Open Caching Working Group for their guidance, contribution, and review.