ALTO Performance Cost Metrics

Cost Metric is a basic concept in Application-Layer Traffic Optimization (ALTO). It is used in both the Cost Map Service and the Endpoint Cost Service. In particular, applications may benefit from knowing network performance measured on several Cost Metrics. For example, a more delay sensitive application may focus on latency, and a more bandwidth-sensitive application may focus on available bandwidth. The objective of this document is to introduce 11 new performance cost metrics, listed in Table 1, to support the aforementioned applications and allow applications to determine "where" to connect based on end to end network performance criteria. Hence, this document extends the base ALTO protocol [ALTO], which defines only a single cost metric, i.e., the generic "routingcost" metric (Sec. 14.2 of ALTO base specification [ALTO]).

+--------+ +--------+ +--------+ | Client | | Client | | Client | +----^---+ +---^----+ +---^----+ | | | +-----------|-----------+ NBI |ALTO protocol | | +--+-----+ retrieve +---------+ | ALTO |<----------------| Routing | | Server | and aggregation| | | |<-------------+ | Protocol| +--------+ | +---------+ | | +---------+ | |Management ---| | | Tool | +---------+ Figure 1.End to End Path Cost Metrics Exposing When an ALTO server supports a cost metric defined in this document, the server SHOULD announce the metric in its IRD. The definitions of a set of cost metrics can allow us to extend the ALTO base protocol (e.g., allowing output and constraints use different cost metrics), but such extensions are not in the scope of this document. One challenge in describing the metrics is that performance metrics often depend on configuration parameters. For example, the value of packet loss rate depends on the measurement interval and varies over time. To handle this issue, ALTO server may collect data on time periods covering the past, present or only collect data on present time. The ALTO may further aggregate these data to provide an abstract and unified view that can be more useful to applications. To make the ALTO client understand whether the performance data is past data or present data, the ALTO server needs to expose to the client the validity period of each performance metric. Following the ALTO base protocol, this document uses JSON to specify the value type of each defined metric. See [RFC4627] for JSON data type specification.

The cost metrics described in this document are similar, in that they may use similar data sources and have similar issues in their calculation. Hence, instead of specifying such issues for each metric individually, we specify the common issue in this section.

An ALTO server needs data sources to compute the cost metrics described in this document. This document does not define the exact data sources. For example, the ALTO server may use log servers or the OAM system as its data source [ALTO-DEPLOYMENT]. In particular, the cost metrics defined in this document can be computed using routing systems as the data sources. Mechanisms defined in [RFC3630], [RFC3784], [OSPF-TE], [ISIS-TE], [BGP-LS] and [BGP-PM] that allow an ALTO Server to retrieve and derive the necessary information to compute the metrics that we described in this document.

An ALTO server processes measurements from data sources to compute exposed metrics. It may need performance data processing tasks such as aggregating the results across multiple systems, removing outliers, and creating additional statistics. One specific challenge in deriving the metrics in this document is that these performance metrics depend on some configuration parameters. For example, the value of packet loss rate depends on the measurement interval and varies over time. If the ALTO server uses aforementioned routing protocol based mechanisms as data sources, then the measurement interval may be preconfigured by the routing protocol. For example, Section 5 of [ISIS-TE] defines a default measurement interval of 30 seconds. This document uses the term Measurement Interval to refer to the measurement interval used by the data sources. In the [ISIS-TE] case, it is a measurement interval set by routing protocol. The Measurement Interval(s) of the data sources can be different from the interval that this document derives the metric, e.g., the interval used by this document is multiple of measurement interval of the data sources. Hence, an ALTO server needs to resolve the mismatch, when it happens. Another issue of converting from data source measurements to ALTO exposed metric values is that the measurement results that the ALTO Server retrieves may be defined for only links, and hence, the server will need to compose the link metrics to obtain path metrics used in services such as the Cost Map Service. In this definition, we define the metrics to be independent of link or path, considering that future ALTO extensions may define link-based services, and hence the defined metrics should still be usable.

Delay To specify spatial and temporal aggregated delay between the specified source and destination or the time that the packet spends to travel from source to destination. The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. The unit is microsecond. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. This is intended to be a constraint attribute value. A Cost Mode is encoded as a US-ASCII string. The Metric value Type is a single 'JSONNumber' type value containing a non-negative integer component that may be followed by an exponent part.This metric could be used as a cost metric constraint attribute used either together with cost metric attribute 'routingcost' or on its own or as a returned cost metric in the response.

Example 1: Delay value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": {"cost-mode" : "numerical", "cost-metric" : "delay"}, "endpoints" : { "srcs": [ "ipv4:192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta" :{ "cost-type": {"cost-mode" : "numerical", "cost-metric" : "delay" } }, "endpoint-cost-map" : { "ipv4:192.0.2.2": { "ipv4:192.0.2.89" : 10, "ipv4:198.51.100.34" : 20, "ipv6:2000::1:2345:6789:abcd" : 30, } } }

Delay jitter To specify spatial and temporal aggregated jitter (latency variation) over the specified source and destination. The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. The unit is microsecond. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 2: Delayjitter value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": {"cost-mode" : "numerical", "cost-metric" : "delayjitter"}, "endpoints" : { "srcs": [ "ipv4:192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost type": { "cost-mode": "numerical", "cost-metric":"delayjitter" } }, "endpoint-cost-map": { "ipv4:192.0.2.2": { "ipv4:192.0.2.89" : 0 "ipv4:198.51.100.34" : 1 "ipv6:2000::1:2345:6789:abcd" : 5 } } }

Packet loss To specify spatial and temporal aggregated packet loss over the specified source and destination. The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. The unit is percentile. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 3: pktloss value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": {"cost-mode" : "numerical", "cost-metric" : "pktloss"}, "endpoints" : { "srcs": [ "ipv4:192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost type": { "cost-mode": "numerical", "cost-metric":"pktloss"} } }, "endpoint-cost-map": { "ipv4:192.0.2.2": { "ipv4:192.0.2.89" : 0, "ipv4:198.51.100.34": 1, "ipv6:2000::1:2345:6789:abcd" : 2, } } }

The metric hopcount is mentioned in [ALTO] as an example. This section further clarifies its properties. Hop count To specify the number of hops in the path between the source endpoint and the destination endpoint. The hop count is a basic measurement of distance in a network and can be exposed as Router Hops, IP hops or other hops in direct relation to the routing prtocols originating this information. it might also result from the aggregation of such information. See section 2.2, Computation of metrics. The unit is integer number. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 4: hopcount value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": {"cost-mode" : "numerical", "cost-metric" : "hopcount"}, "endpoints" : { "srcs": [ "ipv4:192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost type": { "cost-mode": "numerical", "cost-metric":"hopcount"} } }, "endpoint-cost-map": { "ipv4:192.0.2.2": { "ipv4:192.0.2.89" : 5, "ipv4:198.51.100.34": 3, "ipv6:2000::1:2345:6789:abcd" : 2, } } }

Bandwidth To specify spatial and temporal aggregated bandwidth over the specified source and destination. The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endhost to endhost); and the temporal unit is specified as the measurement interval in the query context. This is just a definition of a class of cost metric 'bandwidth'. The use of this cost metric is always in conjunction with what it represents, which could be Max Bandwidth (maxbw), Residual Bandwidth (residuebw) etc. See section 2.2, Computation of metrics. The units are bytes per second. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Maximum Bandwidth To specify spatial and temporal maximum bandwidth over the specified source and destination. The values correspond to the maximum bandwidth that can be used (motivated from RFC 3630 Sec. 2.5.6.). The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endhost to endhost); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition for the Bandwidth Cost Metric. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 5: maxbw value on source-destination endpoint pairs POST/ endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": { "cost-mode": "numerical", "cost-metric": "maxbw"}, "endpoints": { "srcs": [ "ipv4 : 192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost-type": { "cost-mode": "numerical", "cost-metric": "maxbw" } }, "endpoint-cost-map": { "ipv4:192.0.2.2": { "ipv4:192.0.2.89": 0, "ipv4:198.51.100.34" : 2000, "ipv6:2000::1:2345:6789:abcd": 5000, } } }

Maximum Reservable Bandwidth To specify spatial and temporal maximum reservable bandwidth over the specified source and destination. The value is corresponding to the maximum bandwidth that can be reserved (motivated from RFC 3630 Sec. 2.5.7.). The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition of the Bandwidth Cost Metric. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 6: maxresbw value on source-destination endpoint pairs POST/ endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type" { "cost-mode": "numerical", "cost-metric": "maxresbw"}, "endpoints": { "srcs": [ "ipv4 : 192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost-type": { "cost-mode": "numerical", "cost-metric": "maxresbw" } }, " endpoint-cost-map": { "ipv4:192.0.2.2" { "ipv4:192.0.2.89" : 0, "ipv4:198.51.100.34": 2000, "ipv6:2000::1:2345:6789:abcd": 5000, } } }

Unreserved Bandwidth To specify spatial and temporal unreserved bandwidth over the specified source and destination. The values correspond to the bandwidth that can be reserved with a setup priority of 0 through 7. Therefore this metric is endcoded as an array of 8 values. The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition for the bandwidth Cost Metric. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 7: unresbw value on source-destination endpoint pairs In this example, the Collection method specifies that the 'unresbw' values are defined as the 'unavailable bandwidth' specified in section 2.5.8 of RFC3630: 8 unavailable bandwidth value are reported in the same OSPF message using the same TLV. Each value is corresponding to the bandwidth that can be reserved with a setup priority of 0 through 7. POST/ endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type" { "cost-mode": "numerical", "cost-metric": "unresbw[1,8]" }, "endpoints": { "srcs": [ "ipv4:192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost-type": { "cost-mode": "numerical", "cost-metric": "unresbw[1,8]" } }, "endpoint-cost-map" { "ipv4:192.0.2.2" { "ipv4:192.0.2.89" : [0,0,0,0,0,0,0,0], "ipv4:198.51.100.34": [0,0,0,0,0,0,0,2000], "ipv6:2000::1:2345:6789:abcd": [0,0,0,0,0,0,0,5000], } } }

Residue Bandwidth To specify spatial and temporal residual bandwidth over the specified source and destination. The value is calculated by subtracting tunnel reservations from Maximum Bandwidth (motivated from [I-D. ietf-isis-te-metric-extensions], Sec.4.5.). The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition of the general Bandwidth. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 8: residuebw value on source-destination endpoint pairs POST/ endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": { "cost-mode": "numerical", "cost-metric": "residubw"}, "endpoints": { "srcs": [ "ipv4 : 192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost-type" { "cost-mode": "numerical", "cost-metric": "residubw" } }, "endpoint-cost-map" { "ipv4:192.0.2.2" { "ipv4:192.0.2.89" : 0, "ipv4:198.51.100.34": 2000, "ipv6:2000::1:2345:6789:abcd": 5000, } } }

Available Bandwidth To specify spatial and temporal availaible bandwidth over the specified source and destination. The value is calculated by subtracting the measured bandwidth used for the actual forwarding of best effort traffic from Residue Bandwidth (motivated from [I-D. ietf-isis-te-metric-extensions], Sec.4.6.). The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition of the general Bandwidth. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 9: availbw value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": { "cost-mode": "numeric", "cost-metric": "availbw"}, "endpoints": { "srcs": [ "ipv4 : 192.0.2.2" ], "dsts": [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost-type": { "cost-mode": "numeric", "cost-metric": "availbw" } }, "endpoint-cost-map": { "ipv4:192.0.2.2" { "ipv4:192.0.2.89" : [6,5,7,8,4,10,7,6], "ipv4:198.51.100.34" : [7,4,6,8,5,9,6,7], "ipv6:2000::1:2345:6789:abcd" : [7,6,8,5,7,9,6,8], } } }

Utilized Bandwidth To specify spatial and temporal utilized bandwidth over the specified source and destination. The value is corresponding to the actual measured bandwidth used for all traffic (motivated from [I-D. ietf-isis-te-metric-extensions], Sec.4.7.). The spatial aggregation unit is specified in the query context (e.g., PID to PID, or endpoint to endpoint); and the temporal unit is specified as the measurement interval in the query context. See section 2.2, Computation of metrics. See definition of the general Bandwidth. See section 2.1, Data sources. See section 2.1, second paragraph for Measurement Timing. See section 3 for use and application.

Example 10: utilbw value on source-destination endpoint pairs POST /endpointcost/lookup HTTP/1.1 Host: alto.example.com Content-Length: TBA Content-Type: application/alto-endpointcostparams+json Accept: application/alto-endpointcost+json,application/alto-error+json { "cost-type": {"cost-mode" : "numerical", "cost-metric" : "utilbw"}, "endpoints": { "srcs" : [ "ipv4 : 192.0.2.2" ], "dsts" : [ "ipv4:192.0.2.89", "ipv4:198.51.100.34", "ipv6:2000::1:2345:6789:abcd" ] } } HTTP/1.1 200 OK Content-Length: TBA Content-Type: application/alto-endpointcost+json { "meta": { "cost type": { "cost-mode": "numerical", "cost-metric": "utilbw" } }, "endpoint-cost-map": { "ipv4:192.0.2.2" { "ipv4:192.0.2.89" : 0, "ipv4:198.51.100.34" : 2000, "ipv6:2000::1:2345:6789:abcd" : 5000, } } }

The properties defined in this document present no security considerations beyond those in Section 15 of the base ALTO specification [ALTO]. However concerns addressed in Sections "15.1 Authenticity and Integrity of ALTO Information", "15.2 Potential Undesirable Guidance from Authenticated ALTO Information" and "15.3 Confidentiality of ALTO Information" remain of utmost importance. Indeed, TE performance is a highly sensitive ISP information and sharing TE metric values in numerical mode requires full mutual confidence between the entities managing the ALTO Server and Client. Numerical TE performance information will most likely be distributed by ALTO Servers to Clients under strict and formal mutual trust agreements. One the other hand, ALTO Clients must be cognizant on the risks attached to such information that they would have acquired outside formal conditions of mutual trust.

IANA has added the following entries to the ALTO cost map Properties registry, defined in Section 3 of [RFCXXX].