Network Working Group Yunbin Xu
Internet Draft CATR
Intended status: Informational
Expires: April, 25, 2015
Guoying Zhang
CATR
Weiqiang Cheng
CMCC
Haomian Zheng
Huawei
October 25, 2014
Use Cases and Requirements of Dynamic Service Control based on
Performance Monitoring in ACTN Architecture
draft-xu-actn-perf-dynamic-service-control-02.txt
Status of this Memo
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six
months and may be updated, replaced, or obsoleted by other documents
at any time. It is inappropriate to use Internet-Drafts as
reference material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed
at http://www.ietf.org/ietf/1id-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed
at http://www.ietf.org/shadow.html
This Internet-Draft will expire on April 25, 2015.
Copyright Notice
Copyright (c) 2014 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info)in effect on the date of
<Xu> Expires April, 2015 [Page 1]
Internet-Draft actn-perf-dynamic-service-control October 2014
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with
respect to this document. Code Components extracted from this
document must include Simplified BSD License text as described in
Section 4.e of the Trust Legal Provisions and are provided without
warranty as described in the Simplified BSD License.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with
respect to this document.
Abstract
This document introduces the dynamic creation, modification and
optimization of services based on the performance monitoring in the
Abstraction and Control of Transport Networks (ACTN) architecture.
Table of Contents
1. Introduction ....................................... 3
2. Use Cases and Requirements for Dynamic Service Control based on
Performance Monitoring .................................. 3
2.1. Dynamic Service Control based on Traffic Monitoring..... 3
2.2. Dynamic Service Control based on SLA monitoring........ 4
3. Workflows of ACTN Control Modules ....................... 5
3.1. Workflows for Traffic Monitoring based Dynamic Service
Control ........................................... 5
3.2. Workflows for SLA monitoring based Dynamic Service control6
4. Requirement for ACTN Interface ......................... 8
4.1. Interface Requirements for Dynamic Service Control Based on
Traffic Monitoring ................................... 8
4.2. Interface Requirements of Dynamic Service Control based on
SLA monitoring ...................................... 9
4.3. Discussion .................................... 10
5. Security Considerations .............................. 10
6. IANA Considerations ................................. 10
7. References ........................................ 10
7.1. Informative References........................... 10
<Xu> Expires April 2015 [Page 2]
Internet-Draft actn-perf-dynamic-service-control October 2014
1. Introduction
The rapid growth of Internet traffic and the emerging applications
such as cloud computing, datacenter interconnection, IP and optical
integration, LTE backhauling, are driving the transport network to
provide dynamic service provisioning based on the customer
requirement and high quality services with guaranteed performance.
For datacenter interconnection services, IP network transit links,
LTE backhauling services or some business customer services, the
traffic vary over time. However, traditional optical network could
only provide connection based on the maximum bandwidth needed. Based
on flow traffic monitoring, it is possible to adjust the connection
bandwidth according to the real bandwidth needed, create new
connections or increase bandwidth when network traffic exceeds some
certain threshold or reduce connection bandwidth when traffic drops
down, thus helping the customers to save cost.
On the other hand, customers have different SLA requirements. Some
customers such as financial service companies need ultra-low-latency
transmission, some other customers has strict requirements on bit
error rate (BER). In order to provide high quality services
according to customer SLA, network provider needs to measure the
service performance, and dynamically provision and optimize services
based on the performance monitoring result.
The optical transport networks support various performance
monitoring mechanisms, such as traffic flow statistics, packet
delay, delay variation, throughput and packet-loss rate for MPLS-TP
and packet OTN networks, BER, FEC error correction counters for OTN
and DWDM networks, etc. These mechanisms can be used to support
dynamic service control based on performance monitoring.
The Abstraction and Control of Transport Networks (ACTN) described
in [ACTN-FWK] provides a centralized control architecture and open
interfaces that can transmit the customer requirements and policies
to the network, and provide customers with the network status to
make a decision. This draft mainly discusses the use cases and
requirements of dynamic service control based on performance
monitoring in ACTN architecture, the requirements for southbound and
northbound interface are also discussed.
2. Use Cases and Requirements for Dynamic Service Control based on
Performance Monitoring
2.1. Dynamic Service Control based on Traffic Monitoring
For LTE backhauling based on MPLS-TP packet transport networks(PTN)
or packet OTN, it is required that real time or semi-real time
<Xu> Expires April 2015 [Page 3]
Internet-Draft actn-perf-dynamic-service-control October 2014
traffic monitoring of the network should be conducted so as to
resize or optimize traffic and do load balance. In IP and optical
network integration scenario, the optical network can bypass IP
transit traffic as far as the transit traffic bandwidth is large
enough to occupy the granularity of an ODUk. Network traffic
monitoring is important to facilitate automatic discovery of the
imbalance of network traffic, and initiate the network optimization,
thus helping the network operator or the virtual network service
provider to use the network more efficiently and save CAPEX/OPEX.
For datacenter interconnection or enterprise leased line services,
the traffic may vary over time and the customer want to pay for the
bandwidth they really used. Therefore, it is important to provide
some mechanism to monitor the network traffic, adjust and optimize
the services dynamically to help the customers save expenses.
In order to support these scenarios, the customers or client layer
network controllers need to send traffic monitoring and control
policies to the network, while the transport network should report
the traffic monitoring results and dynamically control and adjust
network connections based on the traffic optimization policy. The
service adjustment or network optimization operations normally
should be initiated with the decision of the customer.
2.2. Dynamic Service Control based on SLA monitoring
Customer services have various SLA requirements, such as service
availability, latency, latency jitter, packet loss rate, BER, etc.
The transport network can satisfy service availability and BER
requirements by providing different protection and restoration
mechanisms. However, for other performance parameters, there are no
such mechanisms.
In order to provide high quality services according to customized
SLAs, one possible solution is to measure the service SLA related
performance parameters, and dynamically optimize service
provisioning based on the performance monitoring results.
When the network performance deterioration that violates the SLA is
detected, service optimization operations such as service rerouting,
creation of new connections could be automatically started.
In order to support this requirement, the customer should be able to
send its SLA information to the network, and the transport network
should determine which performance parameters need to be monitored
and the strategy of service optimization. When the service
performance degradation is detected, the transport network can
<Xu> Expires April 2015 [Page 4]
Internet-Draft actn-perf-dynamic-service-control October 2014
notify the customer and immediately start the service optimization
procedure, so as to reduce the impact on the service.
3. Workflows of ACTN Control Modules
In the ACTN architecture [ACTN-FWK], centralized controllers
including Physical Network Controller (PNC), Virtual Network
Controller (VNC), Customer Network Controller, and interfaces
between them have been defined.
For different use cases and scenarios the workflows across the
Customer Network Controller, VNC and PNC are different.
3.1. Workflows for Traffic Monitoring based Dynamic Service Control
Figure 1 shows the workflows for dynamic service control based on
traffic monitoring.
In order to realize dynamic service creation, adjustment and
optimization based on traffic monitoring, the Customer Network
Controller should send traffic monitoring and traffic optimization
strategies to Virtual Network Controller (VNC). VNC sends the
corresponding path traffic monitoring request to PNC. Traffic
monitoring parameters and monitoring cycle need to be carried in
this request.
PNC gets the traffic monitoring results from the underlying physical
networks, translates the monitoring results of the physical topology
to the performance information of the abstract topology, and then
reports to VNC. According to the traffic optimization strategy
obtained from the Customer Network Controller, VNC determines the
optimization procedure, which may include but not limited to
adjusting an existing service, or creating a new connection for load
balance. If it needs to, then VNC send the traffic monitoring
results to the Customer Network Controller, indicating that the
services that need adjustment.
Customer Network Controller firstly confirms whether the service can
be optimized, and then sends a service adjustment request to VNC.
VNC will convert it into path modification or creation request, and
send it to PNC to complete the service optimization. After that, PNC
returns the optimization results to VNC, and VNC passes the results
to the Customer Network Controller.
+------------------------------------------------+
| Customer +-----------------------------+ |
| Network | Dynamic Service Control APP | |
| Controller +-----------------------------+ |
<Xu> Expires April 2015 [Page 5]
Internet-Draft actn-perf-dynamic-service-control October 2014
+------------------------------------------------+
1.Traffic| /|\4.Traffic | /|\
Monitor& | | Monitor | | 8.Traffic
Optimize | | Result 5.Service | | modify &
Policy | | modify& | | optimize
\|/ | optimize Req.\|/ | result
+------------------------------------------------+
| Virtual +-------------------------------+ |
| Network |Dynamic Service Control Agent | |
| Controller +-------------------------------+ |
| +---------------+ +-------------------+ |
| | Flow Optimize | | vConnection Agent | |
| +---------------+ +-------------------+ |
+------------------------------------------------+
2. Path | /|\3.Traffic | |
Monitor | | Monitor | |7.Path
Request | | Result 6.Path | | modify &
| | modify& | | optimize
\|/ | optimize Req.\|/ | result
+--------------------------------------------------------------+
| Physical +----------------------+ +----------------------+ |
| Network | Network Provisioning | |Abstract Topology Gen.| |
| Controller +----------------------+ +----------------------+ |
| +------------------+ +--------------------+ |
| |Network Monitoring| |Physical Topology DB| |
| +------------------+ +--------------------+ |
+--------------------------------------------------------------+
Figure 1 Workflows for dynamic service control based on traffic
monitoring
3.2. Workflows for SLA monitoring based Dynamic Service control
Figure 2 shows the workflows for dynamic service control based on
SLA related performance monitoring.
Customer Network Controller sends the customer service SLA
information and the performance based optimization strategy to VNC.
VNC will convert the SLA information to path performance monitoring
request, which carries the performance monitoring parameters such as
delay, jitter, packet loss, bit error rate and monitoring cycle, and
then send it to the PNC.
<Xu> Expires April 2015 [Page 6]
Internet-Draft actn-perf-dynamic-service-control October 2014
PNC starts the performance monitoring in the underlying physical
networks, collects the results of related path, translates the
performance results of the physical topology to the performance
information of the abstract topology, and reports to VNC. VNC
determines whether the relevant performance parameters can satisfy
the SLA agreements. If the performance degradation seriously
influences the service, for example the service packet delay exceeds
the performance threshold, the VNC will immediately start the
optimization procedure for adjustment. Then the performance
monitoring results as well as the optimizing or adjusting results
will be send to the Customer Network Controller.
+------------------------------------------------+
| Customer +-----------------------------+ |
| Controller | Dynamic Service Control APP | |
| +-----------------------------+ |
+------------------------------------------------+
1. SLA& | /|\6.SLA |
Optimize | | Monitor, modify & |
Policy | | Optimize |
| | Result | 7.Ack
\|/ | \|/
+------------------------------------------------+
| Virtual +-------------------------------+ |
| Network |Dynamic Service Control Agent | |
| Controller +-------------------------------+ |
| +---------------+ +-------------------+ |
| | Flow Optimize | | vConnection Agent | |
| +---------------+ +-------------------+ |
+------------------------------------------------+
2. Path | /|\3.SLA | /|\
Monitor | | Monitor | |5.Path
Request | | Result 4.Path | | Modify &
| | Modify& | | Optimize
\|/ | Optimize Req.\|/ | Result
+--------------------------------------------------------------+
| Physical +----------------------+ +----------------------+ |
| Network | Network Provisioning | |Abstract Topology Gen.| |
| Controller +----------------------+ +----------------------+ |
| +------------------+ +--------------------+ |
| |Network Monitoring| |Physical Topology DB| |
| +------------------+ +--------------------+ |
+--------------------------------------------------------------+
<Xu> Expires April 2015 [Page 7]
Internet-Draft actn-perf-dynamic-service-control October 2014
Figure 2 Workflows for dynamic service control based on SLA
monitoring
4. Requirement for ACTN Interface
ACTN Interfaces defined in [ACTN-FWK] includes the following:
o Consumer-VNC Interface (CVI): an interface between a Customer
Network Controller and a virtual network controller.
o VNC-PNC Interface (VPI): an interface between a virtual network
controller and a physical network controller.
4.1. Interface Requirements for Dynamic Service Control Based on
Traffic Monitoring
According to the work flow of dynamic service control based on
performance monitoring, the information carried in CVI interface
mainly relates to the traffic monitoring and control strategy, while
the VPI interface mainly relates to transports path related traffic
monitoring parameters and results.
1. CVI Interface
The following information used by the Customer Network Controller
should besent to VNC through the CVI.
o Customer service performance monitoring strategy, including the
traffic monitoring object (the service need to be monitored),
monitoring parameters (e.g., transmitted and received bytes per
unit time), traffic monitoring cycle (e.g., 15 minutes, 24 hours),
threshold of traffic monitoring (e.g., high and low threshold),
etc.
o Customer service optimization strategy, such as enabling service
creation or modification when traffic exceeds the threshold.
The following information used by VNC should be sent to the Customer
Network Controller through VPI.
o Traffic monitoring results, to indicate if the traffic exceeds
the bandwidth threshold.
2. VPI Interface
The following parameters used by the VNC should be sent to PNC
through VPI.
<Xu> Expires April 2015 [Page 8]
Internet-Draft actn-perf-dynamic-service-control October 2014
o Traffic monitoring parameters, monitoring object, monitoring cycle,
performance threshold.
The following information used by the PNC should be sent to VNC
through VPI.
o Traffic monitoring results. These results must be translated from
the physical topology to abstract topology by the Abstract
Topology Generalization module firstly.
4.2. Interface Requirements of Dynamic Service Control based on SLA
monitoring
According to the work flow of dynamic service control based on SLA
monitoring, the information in VCI interface mainly contains the SLA
related information and measurement strategy, while the VPI
interface mainly transports path related performance monitoring
parameters and results.
1. CVI Interface
The following information used by the Customer Network Controller
should be sent to the VNC through CVI.
o SLA related performance requirement information, including the
required quality of service parameters (e.g., BER, delay, delay
jitter, packet loss rate, throughput, etc.).
o Service optimization strategy, including the service performance
degradation thresholds and the consequent operations that are
allowed (e.g., rerouting).
The following information is used by the Customer Network Controller
to send to VNC through CVI.
o Monitoring results of service performance, including performance
monitoring parameters, and the services that have been influenced.
o Service optimization results based on performance.
2. VPI Interface
The following information used by VNC should be sent to PNC through
VPI.
o The path performance monitoring request parameters, monitoring
cycle and threshold.
The following information of PNC should be sent to VNC via VPI.
<Xu> Expires April 2015 [Page 9]
Internet-Draft actn-perf-dynamic-service-control October 2014
o Path performance monitoring results.
4.3. Discussion
Performance monitoring in a large scale network could generate a
huge amount of performance information. Therefore, the appropriate
way to deliver the information in CVI and VPI interfaces should be
carefully considered.
5. Security Considerations
This document raises no new security issues.
6. IANA Considerations
No new IANA considerations are raised by this document.
7. References
7.1. Informative References
[ACTN-FWK] Daniele C., Luyuan Fang, Yong Lee and Diego Lopez,
"Framework for Abstraction and Control of Transport
Networks",draft-ceccarelli-actn-framework-03.
Authors' Address
Yunbin Xu
China Academy of Telecom Research
NO.52 Huayuan Beilu, Haidian District, Beijing, China
Email: xuyunbin@catr.cn
Guoying Zhang
China Academy of Telecom Research
NO.52 Huayuan Beilu, Haidian District, Beijing, China
Email: zhangguoying@catr.cn
Weiqiang Cheng
China Mobile Communication Company
Email:chengweiqiang@chinamobile.com
Haomian Zheng
Huawei Technologies
F3-1-B R&D Center, Bantian, Longgang District Shenzhen, China
Email: zhenghaomian@huawei.com
<Xu> Expires April 2015 [Page 10]
Internet-Draft actn-perf-dynamic-service-control October 2014
<Xu> Expires April 2015 [Page 11]