ETSI GS NFV-TST 008-2017 Network Functions Virtualisation (NFV) Release 2 Testing NFVI Compute and Network Metrics Specification (V2 1 1).pdf

1、 ETSI GS NFV-TST 008 V2.1.1 (2017-05) Network Functions Virtualisation (NFV) Release 2; Testing; NFVI Compute and Network Metrics Specification Disclaimer The present document has been produced and approved by the Network Functions Virtualisation (NFV) ETSI Industry Specification Group (ISG) and rep

2、resents the views of those members who participated in this ISG. It does not necessarily represent the views of the entire ETSI membership. GROUP SPECIFICATION ETSI ETSI GS NFV-TST 008 V2.1.1 (2017-05)2 Reference DGS/NFV-TST008 Keywords network, NFV, NFVI, SDN, testing ETSI 650 Route des Lucioles F-

3、06921 Sophia Antipolis Cedex - FRANCE Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16 Siret N 348 623 562 00017 - NAF 742 C Association but non lucratif enregistre la Sous-Prfecture de Grasse (06) N 7803/88 Important notice The present document can be downloaded from: http:/www.etsi.org/standards-sea

4、rch The present document may be made available in electronic versions and/or in print. The content of any electronic and/or print versions of the present document shall not be modified without the prior written authorization of ETSI. In case of any existing or perceived difference in contents betwee

5、n such versions and/or in print, the only prevailing document is the print of the Portable Document Format (PDF) version kept on a specific network drive within ETSI Secretariat. Users of the present document should be aware that the document may be subject to revision or change of status. Informati

6、on on the current status of this and other ETSI documents is available at https:/portal.etsi.org/TB/ETSIDeliverableStatus.aspx If you find errors in the present document, please send your comment to one of the following services: https:/portal.etsi.org/People/CommiteeSupportStaff.aspx Copyright Noti

7、fication No part may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm except as authorized by written permission of ETSI. The content of the PDF version shall not be modified without the written authorization of ETSI. The copyright

8、 and the foregoing restriction extend to reproduction in all media. European Telecommunications Standards Institute 2017. All rights reserved. DECTTM, PLUGTESTSTM, UMTSTMand the ETSI logo are Trade Marks of ETSI registered for the benefit of its Members. 3GPPTM and LTE are Trade Marks of ETSI regist

9、ered for the benefit of its Members and of the 3GPP Organizational Partners. oneM2M logo is protected for the benefit of its Members GSM and the GSM logo are Trade Marks registered and owned by the GSM Association. ETSI ETSI GS NFV-TST 008 V2.1.1 (2017-05)3 Contents Intellectual Property Rights 4g3F

10、oreword . 4g3Modal verbs terminology 4g3Introduction 4g31 Scope 5g32 References 5g32.1 Normative references . 5g32.2 Informative references 5g33 Definitions and abbreviations . 6g33.1 Definitions 6g33.2 Abbreviations . 6g34 Time and Time Intervals for Metrics 6g35 Framework for Metric Definitions .

11、6g36 Compute Metrics 7g36.1 Background 7g36.2 Name 7g36.3 Parameters 8g36.4 Scope 8g36.5 Units of Measure 8g36.6 Definition . 8g36.7 Method of Measurement 8g36.8 Sources of Error . 9g36.9 Discussion 9g37 Network Metrics . 9g37.1 Background 9g37.2 Name 9g37.3 Parameters 10g37.4 Scope 10g37.5 Units of

12、 Measure 10g37.6 Definition . 10g37.7 Method of Measurement 11g37.8 Sources of Error . 11g37.9 Discussion 11g38 Memory Metrics . 11g38.1 Background 11g38.2 Name 11g38.3 Parameters 12g38.4 Scope 12g38.5 Units of Measure 12g38.6 Definition . 12g38.7 Method of Measurement 12g38.8 Sources of Error . 12g

13、38.9 Discussion 13g39 Follow-on Activities . 13g3Annex A (informative): Examples of Scope Specification for Metrics 14g3A.1 Description . 14g3Annex B (informative): Authors Essential, or potentially Essential, IPRs notified to ETSI in respect of ETSI standards“, which is available from the ETSI Secr

14、etariat. Latest updates are available on the ETSI Web server (https:/ipr.etsi.org/). Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates

15、on the ETSI Web server) which are, or may be, or may become, essential to the present document. Foreword This Group Specification (GS) has been produced by ETSI Industry Specification Group (ISG) Network Functions Virtualisation (NFV). Modal verbs terminology In the present document “shall“, “shall

16、not“, “should“, “should not“, “may“, “need not“, “will“, “will not“, “can“ and “cannot“ are to be interpreted as described in clause 3.2 of the ETSI Drafting Rules (Verbal forms for the expression of provisions). “must“ and “must not“ are NOT allowed in ETSI deliverables except when used in direct c

17、itation. Introduction Although many metrics for the performance and utilization of the Network Function Virtualisation Infrastructure (NFVI) components have been in wide use for many years, there were no independent specifications to support consistent metric development and interpretation. The pres

18、ent document provides the needed specifications for key NFVI metrics. ETSI ETSI GS NFV-TST 008 V2.1.1 (2017-05)5 1 Scope The present document specifies detailed and vendor-agnostic key operational performance metrics at different layers of the NFVI, especially processor usage and network interface u

19、sage metrics. These metrics are expected to serve as references for processed and time-aggregated measurement values for performance management information that traverses the Or-Vi and/or Vi-Vnfm reference points of the NFV architectural framework. The present document contains normative provisions.

20、 2 References 2.1 Normative references References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (incl

21、uding any amendments) applies. Referenced documents which are not found to be publicly available in the expected location might be found at https:/docbox.etsi.org/Reference. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long ter

22、m validity. The following referenced documents are necessary for the application of the present document. Not applicable. 2.2 Informative references References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, on

23、ly the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long term validity. The following referen

24、ced documents are not necessary for the application of the present document but they assist the user with regard to a particular subject area. i.1 ETSI GS NFV-INF 003 (V.1.1.1) (12-2014): “Network Functions Virtualisation (NFV); Infrastructure; Compute Domain“. i.2 Linux/UNIX system programming trai

25、ning, Linux man-pages: “TOP(1)“. NOTE: Available at http:/man7.org/linux/man-pages/man1/top.1.html#2._SUMMARY%C2%A0Display. i.3 OReilly Linux Dev Center: “Exploring the /proc/net/ Directory“. NOTE: Available at http:/ i.4 RHEL 6.8 Deployment Guide: “E.2.18. /proc/meminfo“. NOTE: Available at https:/

26、 i.5 ETSI GS NFV 003: “Network Functions Virtualisation (NFV); Terminology for Main Concepts in NFV“. i.6 IETF RFC 7348: “Virtual eXtensible Local Area Network (VXLAN): A Framework for Overlaying Virtualized Layer 2 Networks over Layer 3 Networks“. ETSI ETSI GS NFV-TST 008 V2.1.1 (2017-05)6 3 Defini

27、tions and abbreviations 3.1 Definitions For the purposes of the present document, the terms and definitions given in ETSI GS NFV 003 i.5 apply. 3.2 Abbreviations For the purposes of the present document, the abbreviations given in ETSI GS NFV 003 i.5 and the following apply: CPU Central Processing U

28、nit CRC Cyclic Redundancy Check HZ Hertz of the system clock, an operating system parameter KiB Kibibytes NFVI Network Function Virtualisation Infrastructure OS Operating System RAM Random Access Memory VIM Virtual Infrastructure Manager VXLAN Virtual eXtensible Local Area Network (IETF RFC 7348 i.6

29、). 4 Time and Time Intervals for Metrics Coherent compute domains i.1 usually need access to a clock with accurate time-of-day (or simply time) and sources of periodic interrupts. Time sources are accessed to provide timestamps for events and log entries that document the recent history of the compu

30、te environment. Periodic interrupts provide a trigger to increment counters and read current conditions in the compute and networking environments. With the foundation of time, date, and periodic interrupts, a measurement system can determine the beginning and end of time intervals, which is a funda

31、mental aspect of metrics that involve counting and collecting events. Table 4-1 specifies requirements applicable to time, date, and periodic interrupts. Table 4-1: Requirements applicable to time, date and periodic interrupts General-Time-01 The compute domain shall be able to take readings from (o

32、r access) a clock with accurate time-of-day and calendar date. General-Time-02 The compute domain shall have a source of periodic interrupts available which are derived from the time-of-day clock, with configurable period (a parameter of metrics that use this feature). When the results from measurem

33、ent systems are collected and reported by management systems, the management systems may provide an additional time and date reading associated with the operation to collect the results, using their own time source. 5 Framework for Metric Definitions The metric definitions in the present document ar

34、e primarily based on the fact that the resources of the NFVI have utilization and performance characteristics that can be assessed by measurement processes. The resources may be implemented in hardware, software (such as virtual resources), or a combination of both. The measurement processes are pri

35、marily implemented in software (such as in the kernel or user space), but may be assisted by features of the hardware. The measured NFVI resources and the measurement processes shall be completely specified in the dimensions of model numbers, firmware versions, software versions, and any other aspec

36、ts that influence the results (such as physical location of the components within a datacentres racks and shelves). For example, the fixed frequency of the physical CPU clock in Hz, which governs the rate that the CPU executes instructions, is one important descriptor of the NFVI. ETSI ETSI GS NFV-T

37、ST 008 V2.1.1 (2017-05)7 Clock Speed may depend on other CPU settings, such as energy-saving power control. For one list of NFVI platform descriptors, see clause 5.1 of ETSI GS NFV-INF 003 i.1. For each metric it specifies, the present document provides the following elements: Background Name Parame

38、ters (input factors) Scope of coverage Unit(s) of measure Definition Method of Measurement Sources of Error Discussion NOTE: The present document specifies well-known metrics, and assumes that Virtual Infrastructure Managers (VIM) will control and expose the metrics as specified here, or will be enh

39、anced to collect and convey the metrics with the required framework elements, which are Name, Parameters, Scope, Units of measure, and the source of the measurement (where the metric was measured, which may be synonymous with the Scope). 6 Compute Metrics 6.1 Background The Central Processing Unit (

40、CPU) is an essential component of every coherent compute domain. Each CPU is a limited resource in terms of the instructions per second it can execute. It is valuable to monitor the utilization of the CPU resource to fulfil the goals of maintaining continued and efficient operations, and for trouble

41、shooting abnormal behaviour to find root causes. For many uses, it is helpful to categorize the CPUs execution time into multiple execution contexts, such as system and user contexts. A compute node may include additional processors beyond the main CPU; the metrics specified in this clause can also

42、be used to measure and report the usage of such processors. VNFs also have a view of CPU resources in terms of execution time they have used during a measurement interval. However, the configured instantiation of the VNF determines how to map its view of virtual processor resource usage to actual ha

43、rdware CPU resources available and used. For example, a VNFs processes may be pinned to one or more CPU cores, or the VNF may be sharing the resources of many CPU cores with other VNFs. 6.2 Name There are two variants of this metric: Processor Usage Processor Utilization The two variants allow repor

44、ting this metric as a percentage. The metric is a function of the scope, set of reported contexts, measurement interval and other factors. ETSI ETSI GS NFV-TST 008 V2.1.1 (2017-05)8 6.3 Parameters The following parameters shall be supported for this metric: Tick interval: the period of timed interru

45、pts when the processors execution context can be recorded. Note that this parameter is an integral part of the method of measurement. The tick interval is sometimes called a “jiffy“. The tick interval is controlled by a system parameter, “HZ“, whose default value shall be 100 for measurements comply

46、ing with the present document. Set of execution contexts: the desired set of processor states with reported utilization. For example, the simplest set includes two states: active and idle. Time in the active context can be calculated as a sum of states with more specific definitions, such as the act

47、ive states user and system. A commonly chosen set of four states is user, system, wait, and idle. See i.2 for a list of seven states available in Linux OS. End time: the termination of the measurement interval (time and date). Measurement interval: the duration of the observation by the measurement

48、system to assess the metric. 6.4 Scope The list of one or more compute resources which shall be included in the values reported, and whether the resource is physical or virtual. Annex A gives examples of the scope usage for physical and virtual processor metrics. 6.5 Units of Measure Processor usage

49、 results shall be reported as time in seconds, and utilization shall be reported as the ratio of total time in one or more execution contexts to the total time in the measurement interval, expressed as a percentage. 6.6 Definition The compute resource usage at time T (for a given scope, context, and measurement interval) shall be assessed as indicated in