ALTO for Querying LMAP ResultsChina TelecomBeijingChinaxiechf@chinatelecom.cnChina Telecom32 Xuanwumen West St, Xicheng DistrictBeijing102209wangw36@chinatelecom.cnHuawei101 Software Avenue, Yuhua DistrictNanjingJiangsu210012Chinamaqiufang1@huawei.com
Transport
Application-Layer Traffic OptimizationALTO LMAPMeasuring broadband performance on a large scale for network
diagnostics is important to providers and users, as well as for public
policy. The Large-scale Measurement of Broadband Performance (LMAP)
framework, information model, and protocol have been developed for
measurement task dissemination, initialization, reporting and
storing.This document uses the ALTO protocol to provide access to large-scale
network measurement results, which could be useful to constitute the
ALTO cost map service and the endpoint cost service. Potential ALTO
protocol extensions are also discussed to better leverage LMAP
measurement results.Measuring broadband performance on a large scale for network
diagnostics is important to providers and users, as well as for public
policy. A framework for Large-scale Measurement of Broadband Performance
(LMAP) has been developed to coordinate the
execution of broadband measurements and the collection of measurement
results across a large network scale.The LMAP framework defines three basic elements: Measurement
Agents(MAs), Controllers, and Collectors. Measurement Agents (MAs)
initiate the actual measurements, which are called Measurement Tasks.
The controller instructs one or more MAs and communicates the set of
Measurement Tasks an MA should perform and when. The Collector accepts
reports from the MAs with the results from their Measurement Tasks. A
YANG data model has been defined for LMAP
platforms .The Application-Layer Traffic Optimization (ALTO) protocol provides a solution to expose network information to
applications. While the ALTO server can provide an abstract and unified
view to the ALTO client, it remains undefined how the ALTO server can
leverage multiple systems to collection and aggregate network
information.This document tries to bridge the gap by proposing the ALTO protocol
to access large-scale network measurement results in the context of
Large-scale Measurement of Broadband Performance (LMAP) . The measurement result reports could be useful to
support the ALTO cost map service or endpoint cost service. Potential
ALTO protocol extensions are also discussed to better leverage LMAP
measurement results.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 motivate the proposal of ALTO for querying LMAP results, consider
some key use cases defined in :Broadband network maintenance and monitoringA network
operator needs to understand the performance of their networks, the
performance of the suppliers (downstream and upstream networks), the
performance of Internet access services, and the impact that such
performance has on the experience of their customers. Largely, the
processes that ISPs operate (which are based on network measurement)
include Identifying, isolating, and fixing problems, Design and
planning, understanding the quality experienced by customers,
Understanding the impact and operation of new devices and
technology.Broadband performance benchmarkingA
regulator may want to evaluate the performance of the Internet
access services offered by operators.While
each jurisdiction responds to distinct consumer, industry, and
regulatory concerns, much commonality exists in the need to produce
datasets that can be used to compare multiple Internet access
service providers, diverse technical solutions, geographic and
regional distributions, and marketed and provisioned levels and
combinations of broadband Internet access services.Regulators may want to publish performance measures
of different ISPs as background information for end users. They may
also want to track the growth of high-speed broadband deployment, or
to monitor the traffic management practices of Internet
providers.This document addresses how to retrieve aggregated network
performance measurement results for a certain network. These network
performance measurement results are measured and gathered using the LMAP
based measurement system. The LMAP based measurement system is comprised
of three components: Measurement Agent (MA),Collector and Controller.
The MA is located in both the ingress node and the egress node and
instructed by the Controller to monitor a particular traffic flowing
toward a given destination and to send the Report to the Collector. The
Report contains:Date and time when the report was sentAgent-id or group-id to identify the Measurement Agent (group)
from which the report originatesthe actual Measurement Results, including the measurement task
name, the task-specific parameters which allow extension, the
additional tags, the task start/end time, the result values,
etc.The collector then provides results to the repository in the ALTO
server, formats it as ALTO information, and exposes it to the ALTO
client, see .It must be possible to query for specific, possibly aggregated,
results in a flexible way. Otherwise, entities interested in measurement
results either cannot select the kind of result aggregation they desire,
or must always fetch large amounts of detailed results and process these
huge datasets themselves. The need for a flexible mechanism to query for
dedicated, partial results becomes evident when considering use cases
where a service provider or a process wants to use certain measurement
results in an automated fashion. For instance, consider a video
streaming service provider that wants to know for a given end-user
request the average download speed of the end user's access provider in
the end user's region (e.g. to optimize/parametrize its http adaptive
streaming service). Or consider a website which is interested in
retrieving average connectivity speeds for users depending on access
provider, region, or type of contract (e.g. to be able to adapt web
content on a per-request basis according to such statistics).The ALTO cost calendar defined in RFC 8896 allows an ALTO Server to
provide a sequence of network costs for a given duration of time. It
provides the capability for applications to figure out the best time
to schedule data transfers and also to proactively manage application
traffic given predictable events, such as an expected spike in traffic
due to crowd gathering (concerts, sports, etc.), traffic-intensive
holidays, and network maintenance .ALTO cost calendar defines "time-interval-size" and
"number-of-intervals" as the calendar attributes to specify the time
interval size and the number of intervals provided in the calendar,
specifically. The calendar mode now seems more like a periodic
recurrence, while lack of a more comprehensive expression of calendar
time. For example, an application may want to know the network cost
metric between two specific endpoints for every 15-minute interval
between 12:00 p.m. and 1:00 p.m., Monday through Wednesday. It is
possible for LMAP to return that result by configuring the event that
triggered the execution of the measurement schedule under the
/lmap/schedules subtree. This requires an extension to ALTO cost
calendar to support the exposure to ALTO client.In addition, some ALTO protocol extensions need to be considered.
For example, Additional entity property types such as measurement points or
report measurement points need to be introduced to indicate where
these results are measured and who reports these measurement
resultsAdditional entity property type such as task name or program
name needs to be introduced to express what task is performedAdditional cost metrics need to be introduced to describe what
performance metrics are collected and what their values areComment: Should we expose LMAP details to ALTO clients?Comment from Luis: how PIDs defined for the measurement agents
could correlate with conventional PIDs, i.e., those representing IP
address pools.TBDThis work provides approach to get access to large scale broadband
network performance data and has benefited from the discussions of
large-scale network measurement data retrieval over the years.This document has no requests to IANA.