Network Working Group T. Kim Internet-Draft H. Yu Intended status: Informational C. Jeong Expires: May 4, 2017 Y. Han E. Paik KT October 31, 2016 Benchmarking Methodology for Service Function Chain Performance draft-kim-bmwg-sfc-benchmark-meth-00 Abstract Service Function Chain is the ordered set of service functions such as firewall, Deep Packet Inspection(DPI), virtualized Evolved Packet Core (vEPC), and etc,. Operators make chains with several service functions depending on the service which they have to provide. The chain needs to be evaluated to measure the SLA. This draft describes the benchmarking methodologies for Service Function Chain(SFC) performance and the affecting factors to SFC performance. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119]. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. 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." This Internet-Draft will expire on May 4, 2017. Kim, et al. Expires May 4, 2017 [Page 1] Internet-Draft sfc performance benchmarking October 2016 Copyright Notice Copyright (c) 2016 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 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. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 3 3. Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.1. Test Topology . . . . . . . . . . . . . . . . . . . . . . 3 3.2. Test Traffic . . . . . . . . . . . . . . . . . . . . . . 4 4. Benchmarking Test . . . . . . . . . . . . . . . . . . . . . . 4 4.1. Connectivity . . . . . . . . . . . . . . . . . . . . . . 4 4.2. Performance . . . . . . . . . . . . . . . . . . . . . . . 5 4.2.1. E2E Latency . . . . . . . . . . . . . . . . . . . . . 5 4.2.2. E2E Packet Loss Rate . . . . . . . . . . . . . . . . 5 4.2.3. E2E Bandwidth . . . . . . . . . . . . . . . . . . . . 6 5. Factors affecting the SFC Performance . . . . . . . . . . . . 6 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 8. Normative References . . . . . . . . . . . . . . . . . . . . 7 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 8 1. Introduction Service Function Chain is the ordered set of service functions such as firewall, Deep Packet Inspection(DPI), virtualized Evolved Packet Core (vEPC), and etc,. The service functions include virtualized network functions and physical network functions. As the network infrastructure become virtualized, operators make chains with several service functions depending on the service which they have to provide. The chain needs to be evaluated to measure the SLA. This draft describes the benchmarking methodologies for Service Function Chain(SFC) performance and the influential factors to SFC performance. Kim, et al. Expires May 4, 2017 [Page 2] Internet-Draft sfc performance benchmarking October 2016 2. Definition of Terms The detail explanations of each term are in [RFC 7665] SF Service Function SFC Service Function Chain SFF Service Function Forwarder CLA Classifier PNF Physical Network Function VNF Virtualized Network Function NSH Network Service Header 3. Test Setup This section discusses test topology and the test traffic 3.1. Test Topology +-------------------------------------------------------------+ | Cloud | | +--------+ +--------+ | | | | | | | | | VNF 1 | | VNF 2 | | | | | | | | | +--------+ +--------+ | | | | | | | | | | | | | | +---------+ +-----------------------+ +---------+ | +--------+ | | | | +---+ +---+ | | | | | | | | vHost 1 |----| |CLA| |SFF| Virtual|----| vHost 2 | | | PNF | | | | | +---+ +---+ Switch | | | | | | | +---------+ +-----------------------+ +---------+ | +--------+ +-------------------------------------------------------------+ | | | | | +--------+ +---------------------------------------------------------------------------+ +--------+ | | | +----------+ +--------------------------+ | | | | Host 3 |----| |Classifier| |Service Function Forwarder| |----| Host 4 | | | | +----------+ +--------------------------+ Physical Switch | | | +--------+ +---------------------------------------------------------------------------+ +--------+ Kim, et al. Expires May 4, 2017 [Page 3] Internet-Draft sfc performance benchmarking October 2016 3.2. Test Traffic There are two types of traffic. One is External traffic and the other is Internal traffic. o Internal Traffic : * The traffic flows inside the cloud. A source host and a destination host are inside the same cloud and the SFC is also made in the cloud. Therefore, the SFC does not contain a SF outside the cloud(PNF). (e.g. SFC : vHost1 -> VNF1 -> VNF2 -> vHost2) o External Traffic : * The traffic flows outside the cloud. A source host or destination host can be exists outside the cloud. Therefore, the SFC can contain a SF outside the cloud(PNF) (e.g. SFC : Host3 -> VNF1 -> VNF2 -> PNF-> Host4) The frame sizes of the test traffic SHOULD be multiple sizes as recommended in RFC2544. 4. Benchmarking Test 4.1. Connectivity Objective : The connectivity of each part of SFC and the end to end SFC it self. This test demonstrates the SFC works properly. Procedure: 1. Send the test traffic from source host to destination host 2. Check each SF and links between the SFs 3. Check the test traffic from the source host and the destination host. 4. Among SFs, the test traffic SHOULD flows only selected SF from the source host to the destination host. Kim, et al. Expires May 4, 2017 [Page 4] Internet-Draft sfc performance benchmarking October 2016 4.2. Performance 4.2.1. E2E Latency Objective : This test demonstrates how much time the SFC takes to flow traffic from the source host to the desination host. Latency is the key of some services such as video streaming. Procedure: 1. Check the connectivity of the SFC 2. Send the test traffic from source host to destination host 3. Check the test traffic from the source host and the destination host. Measurement: E2E Latency Time = TL Average E2E Latency : TL1 + TL2 + ...TLn ---------------------- Total Test Iterations 4.2.2. E2E Packet Loss Rate Objective : This test demonstrates how many packets are loss depending on the frame sizes or parallel SFCs Procedure: 1. Check the connectivity of the SFC 2. Make the conflict circumstances with differenct frame sizes and other SFCs 3. Send the test traffic from source host to destination host. 4. Check the test traffic from the source host and the destination host. Kim, et al. Expires May 4, 2017 [Page 5] Internet-Draft sfc performance benchmarking October 2016 Measurement: E2E Packet Loss Rate = PLR Average Packet Loss Rate : PLR1 + PLR2 + ...PLRn ------------------------ Total Test Iterations 4.2.3. E2E Bandwidth Objective : This test demonstrates how much bandwidth the SFC can support. To find out the bandwidth of SFC is enough for particular sevices such as bandwidth-intensive services. Procedure: 1. Check the connectivity of the SFC 2. Send the test traffic from source host to destination host. 3. Check the test traffic from the source host and the destination host has no packet loss. 4. Record the E2E Bandwidth. Measurement: E2E Bandwidth = BW Average E2E Bandwidth : BW1 + BW2 + ...BWn --------------------- Total Test Iterations 5. Factors affecting the SFC Performance This section describes factors affecting the SFC performance. o SFC awareness * - Depending on the awareness of SFC encapsulation,NSH, the SFC performance is different. When SFC uses NSH, it takes time to check the NSH of every packet. Kim, et al. Expires May 4, 2017 [Page 6] Internet-Draft sfc performance benchmarking October 2016 o Composition of SFC * the number of SFs in the SFC affects the SFC performance because of the trasition overhead. o Operation of SF * The operations of SF can affect to the SFC performance, such as DPI and UTM. * When the SF has multi functions, the traffic takes time to pass through the SF. o Types of SF; PNF or VNF * It is hard to assure the network performance of VNF because it is on the virtual machine(VM); VNF is affected from the CPU of physical machine(PM). * VNF is also affected from the number of flow rules in the virtual switch. 6. Security Considerations TBD. 7. IANA Considerations No IANA Action is requested at this time. 8. Normative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for Network Interconnect Devices", RFC 2544, DOI 10.17487/RFC2544, March 1999, . [RFC7665] Halpern, J., Ed. and C. Pignataro, Ed., "Service Function Chaining (SFC) Architecture", RFC 7665, DOI 10.17487/RFC7665, October 2015, . Kim, et al. Expires May 4, 2017 [Page 7] Internet-Draft sfc performance benchmarking October 2016 Authors' Addresses Taekhee Kim KT Infra R&D Lab. KT 17 Woomyeon-dong, Seocho-gu Seoul 137-792 Korea Phone: +82-2-526-6688 Fax: +82-2-526-5200 Email: taekhee.kim@kt.com Hyun Yu KT Infra R&D Lab. KT 17 Woomyeon-dong, Seocho-gu Seoul 137-792 Korea Phone: +82-2-526-6688 Fax: +82-2-526-5200 Email: hyun.yu@kt.com Chiwook Jeong KT Infra R&D Lab. KT 17 Woomyeon-dong, Seocho-gu Seoul 137-792 Korea Phone: +82-2-526-6688 Fax: +82-2-526-5200 Email: chiwook.jeong@kt.com Youngtae Han KT Infra R&D Lab. KT 17 Woomyeon-dong, Seocho-gu Seoul 137-792 Korea Phone: +82-2-526-6688 Fax: +82-2-526-5200 Email: youngtae.han@kt.com Kim, et al. Expires May 4, 2017 [Page 8] Internet-Draft sfc performance benchmarking October 2016 EunKyoung Paik KT Infra R&D Lab. KT 17 Woomyeon-dong, Seocho-gu Seoul 137-792 Korea Phone: +82-2-526-5233 Fax: +82-2-526-5200 Email: eun.paik@kt.com URI: http://mmlab.snu.ac.kr/~eun/ Kim, et al. Expires May 4, 2017 [Page 9]