Probabilistic Loss Ratio Search for Packet Throughput (PLRsearch)
draft-vpolak-bmwg-plrsearch-00

Document Type Active Internet-Draft (individual)
Last updated 2018-11-13
Replaces draft-vpolak-plrsearch
Stream (None)
Intended RFC status (None)
Formats plain text xml pdf html bibtex
Stream Stream state (No stream defined)
Consensus Boilerplate Unknown
RFC Editor Note (None)
IESG IESG state I-D Exists
Telechat date
Responsible AD (None)
Send notices to (None)
Benchmarking Working Group                       M. Konstantynowicz, Ed.
Internet-Draft                                             V. Polak, Ed.
Intended status: Informational                             Cisco Systems
Expires: May 17, 2019                                  November 13, 2018

   Probabilistic Loss Ratio Search for Packet Throughput (PLRsearch)
                     draft-vpolak-bmwg-plrsearch-00

Abstract

   This document addresses challenges while applying methodologies
   described in [RFC2544] to benchmarking NFV (Network Function
   Virtualization) over an extended period of time, sometimes referred
   to as "soak testing".  More specifically to benchmarking software
   based implementations of NFV data planes.  Packet throughput search
   approach proposed by this document assumes that system under test is
   probabilistic in nature, and not deterministic.

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 https://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 17, 2019.

Copyright Notice

   Copyright (c) 2018 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
   (https://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

Konstantynowicz & Polak   Expires May 17, 2019                  [Page 1]
Internet-DraftProbabilistic Loss Ratio Search for Packet ThNovember 2018

   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Motivation  . . . . . . . . . . . . . . . . . . . . . . . . .   2
   2.  Model . . . . . . . . . . . . . . . . . . . . . . . . . . . .   2
   3.  Poisson Distribution  . . . . . . . . . . . . . . . . . . . .   3
   4.  Fitting Function Coefficients Distribution  . . . . . . . . .   4
   5.  Algorithm Formulas  . . . . . . . . . . . . . . . . . . . . .   5
     5.1.  Integration . . . . . . . . . . . . . . . . . . . . . . .   5
     5.2.  Optimizations . . . . . . . . . . . . . . . . . . . . . .   5
   6.  Known Implementations . . . . . . . . . . . . . . . . . . . .   5
     6.1.  FD.io CSIT Implementation Specifics . . . . . . . . . . .   5
   7.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .   5
   8.  Security Considerations . . . . . . . . . . . . . . . . . . .   6
   9.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . .   6
   10. Normative References  . . . . . . . . . . . . . . . . . . . .   6
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .   6

1.  Motivation

   Network providers are interested in throughput a device can sustain.

   RFC 2544 assumes loss ratio is given by a deterministic function of
   offered load.  But NFV software devices are not deterministic
   (enough).  This leads for deterministic algorithms (such as MLRsearch
   with single trial) to return results, which when repeated show
   relatively high standard deviation, thus making it harder to tell
   what "the throughput" actually is.

   We need another algorithm, which takes this indeterminism into
   account.

2.  Model

   Each algorithm searches for an answer to a precisely formulated
   question.  When the question involves indeterministic systems, it has
   to specify probabilities (or prior distributions) which are tied to a
   specific probabilistic model.  Different models will have different
   number (and meaning) of parameters.  Complicated (but more realistic)
   models have many parameters, and the math involved can be very
   complicated.  It is better to start with simpler probabilistic model,
   and only change it when the output of the simpler algorithm is not
   stable or useful enough.

   TODO: Refer to packet forwarding terminology, such as "offered load"
   and "loss ratio".

Konstantynowicz & Polak   Expires May 17, 2019                  [Page 2]
Internet-DraftProbabilistic Loss Ratio Search for Packet ThNovember 2018

   TODO: Mention that no packet duplication is expected (or is filtered
   out).

   TODO: Define critical load and critical region earlier.
Show full document text