DICE: Performance Update

1
DICE Performance Update

• Eric L. Boyd (Internet2)
• Joe Metzger (ESnet)
• Nicolas Simar (G2 JRA1)

2
Vision Performance Information is

• Available
• People can find it (Discovery)
• Community of trust allows access across
  administrative domain boundaries (AA)
• Ubiquitous
• Widely deployed (Paths of interest covered)
• Reliable (Consistently configured correctly)
• Valuable
• Actionable (Analysis suggests course of action)
• Automatable (Applications act on data)

3
Getting There Build Empower the Community

• Decouple the Problem Space
• Analysis and Visualization
• Performance Data Sharing
• Performance Data Generation
• Grow the Footprint
• Clean APIs and protocols between each layer
• Widespread deployment of measurement
  infrastructure
• Widespread deployment of common performance
  measurement tools

4
perfSONAR Credits

• GÉANT2 JRA1 includes
• Arnes
• Belnet
• Carnet
• Cesnet
• DANTE
• DFN
• FCCN
• GRNet
• GARR
• ISTF
• PSNC
• Nordunet (Uninett)
• Renater
• RedIRIS
• Surfnet
• SWITCH

• perfSONAR is a joint effort
• ESnet
• Fermilab
• GÉANT2 JRA1
• Internet2
• RNP
• Internet2 includes
• University of Delaware
• Georgia Tech
• Internet2 staff

5
perfSONAR Project Activity Meter

• Interactions
• 1-2 conf calls/week
• 1 new service/month (accelerating)
• 3-4 development workshops/year
• 3-4 paper submissions/year
• Recruitment
• RNP has joined the effort
• Outreach to LHC community
• GaTech beginning six month commitment

6
perfSONAR Services (1)

• Measurement Point Service
• Enables the initiation of performance tests
• Measurement Archive Service
• Stores performance monitoring results
• Lookup Service
• Allows the client to discover the existing
  services and other LS services.
• Dynamic services registration themselves to the
  LS and mention their capabilities, they can also
  leave or be removed if a service gets down.
• AuthN/Z Service
• Internet2 MAT, GN2-JRA5 (eduGAIN)
• Authorization functionality for the framework
• Users can have several roles, the authorisation
  is done based on the user role.
• Trust relationships defined between users
  affiliated with different administrative domains.

7
perfSONAR Services (2)

• Transformation Service
• Transform the data (aggregation, concatenation,
  correlation, translation, etc).
• Topology Service
• Make the network topology information available
  to the framework.
• Find the closest MP, provide topology information
  for visualisation tools
• Resource protector
• Arbitrate the consumption of limited resources
  between multiple services.

8
Types of perfSONAR Services

• Core Services
• Set released by perfSONAR Team
• e.g. LS, AA, 3 MPs, 2 MAs, RP, Tos, TS
• Tested for interoperability
• Serve as examples for affiliated developers
• Targeted at next generation network needs (e.g.
  GÉANT2, Internet2 New Network, etc.)
• Affiliated Services
• Released by perfSONAR partners, lag Core
• May share development infrastructure (Bugzilla,
  Website, Mailing Lists)
• Candidates for migration to Core Services
• Unaffiliated Services

9
perfSONAR Core Status Update

• Production release of core services package v1.0
  ready (pending licensing completion)
• Core services include
• Single domain LS solution (PSNC)
• RRD MA (PSNC)
• Affiliate services and client applications
  supporting this version will soon follow
• BWCTL MP (DFN)
• perfSONAR UI (ISTF)
• Ongoing work
• AA Design (Internet2, JRA1, JRA5)
• Multi-LS (PSNC, RNP, UDel)
• ToS (DFN, UDel)

10
perfSONAR Process Status Update

• We have processes -)
• Release management process implemented
  (Internet2, RedIRIS, UDel)
• Bugzilla up and running (UDel)
• Migrated from CVS to SVN (Internet2)
• Functional testing under construction (GRnet)
• Monitoring deployed services with Tomcat (ISTF)
• Installation process eased significantly (DANTE,
  PSNC, UDel)
• www.perfsonar.net under development (Internet2,
  Renater)
• Development information will stay on the Wiki
• Adopter information will migrate to website

11
perfSONAR Affiliate Status Update

• Affiliated Services
• Command Line Interface MP (Ping, OWAMP,
  Traceroute) (RNP, released)
• BWCTL MP (DFN, released)
• SQL MA (PSNC, released)
• L2-specific MA (DANTE)
• SSH MP (Looking Glass) (Belnet, released)
• ABW MP (bandwidth packet capture cards) (Cesnet)
• NMS MP (SDH status) (DANTE)
• Hades MA (OWD, Jitter, OWPL) (DFN)
• Flow Replicator MA (Surfnet, Carnet)

• User Interfaces
• CNM (DFN)
• perfSONAR UI (ISTF)
• Visual PerfSONAR (Carnet)
• Looking Glass (Belnet)
• ICE/NeTraMet (RNP)

12
What You See Is What You Get

• perfsonarUI
• Retrieval of published data
• RRD MA
• Hades MA
• Visualisation of OWD, IPDV and packet loss
  between Hades MP
• Parsing of arbitrary IPv4 or IPv6 traceroute
  commands
• CNM map based
• GEANT2 NRENS maps
• VisualperfSONAR
• Looking Glass

13
RRD MA features

• Wrapper around RRD tool.
• Request/reply interface.
• Write into RRD.
• LS registration.
• Installation scripts.
• Test configuration files available.

14
Lookup Service Features

• Centralized LS (Creating a distributed LS is
  ongoing development)
• Service Registration (including updates)
  functionality
• Service deregistration functionality
• Lookup/query functionality (XQuery/XPath)
• Services keep-alives
• including database cleanup, scheduled
  functionality
• Registration component for a service available.
• Installation scripts.

15
RRD MA deployment Status
16
MA deployment over time
17
PerfSONAR Next steps

• Formal partnership
• License, Partnership Agreement
• Interim solution
• Upgrade existing user base (currently using
  prototype)
• Data exchange policy (measurement peering
  agreement)
• Consistent offer of services.
• What services package to suggest to networks.
• L2 status monitoring.

18
ESnet

• Joe Metzger

19
GÉANT2 JRA1
20
Last months (Jan May) - 1

• Services
• Lookup Service
• Centralized
• Registration / deregistration
• Lookup query
• Result code
• SQL MA
• Stores data in relational database
• Supports
• Utilization
• L2 status
• Result code.
• HADES MA
• Provides access to the data archive of Hades
  measurements from GEANT2 network

21
Last months - 2

• Tools integration
• Telnet / SSH MP
• On-demand requests for device specific
  information
• Cisco/Juniper/Quagga support
• Resource protection mechanisms
• To verify the parameters send in the commands
• To prevent flood of requests
• BWCTL / OWAMP MP
• BWCTL
• TCP throughput measurements
• OWAMP
• OWD, PL measurements

22
Last months - 3

• Tools integration
• Passive
• ABW
• Counts the number of captured packets and bytes
  and computes used bandwidth
• Short timescale intervals
• Tracefile Capture Measurement Point (TCMP)
• Used for capturing packets of selected flows
  using either regular Eth cards or special DAG or
  COMBO6 cards
• SNMP MP
• Web Service access to the usage of SNMP
• Get for now and OID discovery

23
Last months - 4

• Alcatel NMS MP
• Web Service access to SDH and WDM monitoring
  parameters such as SES, ES, UAS and also the
  G.709 metric BBE
• Acts as a reference implementation
• Can be used by other NRENs in order to build
  perfSONAR compliant services which can retrieve
  data from NMS
• AA
• Designing and developing a perfSONAR AA service
  making use of JRA5's eduGAIN
• Topology Service
• Common schema with SA3

24
What You See Is What You Get

• perfsonarUI
• Retrieval of published data
• RRD MA
• Hades MA
• Visualisation of OWD, IPDV and packet loss
  between Hades MP
• Parsing of arbitrary IPv4 or IPv6 traceroute
  commands
• CNM map based
• GEANT2 NRENS maps
• VisualperfSONAR
• Looking Glass

25
Meet the NRENs sessions

• Powerful tools and useful information
• Design (MAs, MPs approach is good)
• The number of deployed services is high
• Friendly user interfaces
• Tools bring a motivation for installing services
  for attendees
• Sharing of info between projects is useful

• Need to integrate the tools in a single
  visualisation application.
• There are too few networks nodes running the
  services
• Not enough data available
• Not enough information available about perfSONAR
• Would like to have libraties/APIs
• Requirement for having its network perfSONAR
  enabled.

26
Next disseminations workshops

• SEEREN2 workshop in Heraklion.
• E2E service status services deployment for NRENs
  next week in Muenchen.
• Three more installation workshop planned over the
  next 12 months.

27
Data Exchange for E2E Monitoring Archive
scenario
Connect. Communicate. Collaborate

• NREN in charge of retrieving the data from the
  NMS/DB to analyse them and pass the information
  to a java class.
• About 700-1000 lines of code for GÉANT 15 days.
• JRA1
• Provides the mySQL MA service code
• maintains it.
• Provides the script to write into the DB
• JRA4 in charge of the E2E NOC visualisation.

28
Year 3 Objectives

• Improving the visualisation and tools features
  (NOC, PERT, project)
• Integration of AA.
• Services deployments.
• Going operational (with SA3 WI15).
• Mastering the amount of data.
• L1-L2
• Dissemination workshops for NRENs.

29
Timeline
30
Phases III

• End of November 2006
• Going operational (SA3 WI-15)
• RRD MA
• LS (plus LS registration for the other services)
• SNMP MP
• perfsonarUI
• CNM
• Hades and RIPE TTM MA
• BWCTL MP.
• L2 status MP.
• Novelties
• Netflow Integration
• Topology Service
• VisualperfSONAR
• Multi-LS
• Push interface

31
Phase IV

• End of May 2007
• Going operational (SA3 WI-15)
• Multi-LS
• Topology Service
• Hades MP
• BWCTL MA
• VisualperfSONAR
• Novelties
• First set of services using JRA5 Authentication
  with some Authorization.
• Performance anomaly detection

32
Phase V

• End of November 2007
• Going operational (SA3 WI-15)
• Authentication Service

33
Internet2

• Eric Boyd

34
Vision Performance Information is

• Available
• People can find it (Discovery)
• Community of trust allows access across
  administrative domain boundaries (AA)
• Ubiquitous
• Widely deployed (Paths of interest covered)
• Reliable (Consistently configured correctly)
• Valuable
• Actionable (Analysis suggests course of action)
• Automatable (Applications act on data)

35
Getting There Build Empower the Community

• Decouple the Problem Space
• Analysis and Visualization
• Performance Data Sharing
• Performance Data Generation
• Grow the Footprint
• Clean APIs and protocols between each layer
• Widespread deployment of measurement
  infrastructure
• Widespread deployment of common performance
  measurement tools

36
Result No more mystery

• Increase network awareness
• Set user expectations accurately
• Reduce diagnostic costs
• Performance problems noticed early
• Performance problems addressed efficiently
• Network engineers can see act outside their
  turf
• Transform application design
• Incorporate network intuition into application
  behavior

37
Immediate Game-plan

• Internet2 is leveraged to help provide diagnostic
  information for backbone portion of problem
• Create some diagnostic tools
• Make Abilene data as public as is reasonable
• Work on efforts to more widely make performance
  data available (perfSONAR)
• Contribute to base development
• Integrate our diagnostic tools as good
  example MP/MA services

38
BWCTL (Bandwidth Controller)

• What is it?
• A resource allocation and scheduling daemon for
  arbitration of iperf tests
• Typical Solution
• Run iperf or similar tool on two endpoints and
  hosts on intermediate paths
• Typical road blocks
• Need permissions on all systems involved
• Need to coordinate testing with others
• Need to run software on both sides with specified
  test parameters

39
BWCTL 3-Party Flow Diagram
40
OWAMP One-Way Active Measurement Protocol

• What is it?
• Measures one-way latency 1-way ping
• Control connection used to broker test request
  based upon policy restrictions and available
  resources. (Bandwidth/disk limits)
• Specification
• http//tools.ietf.org/wg/ippm/draft-ietf-ippm-owdp
  /draft-ietf-ippm-owdp-14.txt

41
OWAMP Flow Diagram
42
Thrulay Overview

• Network capacity and delay tester
• Same class of tools as iperf, netperf, nettest,
  nuttcp, ttcp, etc.
• Unique features not found in other tools
• TCP measures round-trip delay along with goodput
• UDP measures
• One-way delay, with quantiles
• Packet loss
• Packet duplication
• Reordering
• UDP ability to send precisely positioned true
  Poisson streams (microsecond errors in sending
  times)
• Human and machine-readable (ready to be fed to
  gnuplot)

43
Thrulay Update

• New release v0.8
• Tests with multiple TCP streams
• Set DSCP (a.k.a. first 6 bits of the TOS byte)
• Report MTU and/or MSS (whichever the OS makes
  available)
• More UDP statistics duplication, reordering,
  quantiles of delay
• SPARC/Solaris support
• Mac OS X support
• IPv6 support
• Non-busy-waiting UDP mode (less precise, but can
  run more concurrent tests)
• Documentation manual pages have been added
• Basic client authorization based on IP address
• Integration of TSC timekeeping projects for
  faster and more precise timestamping

44
NDT Network Diagnostic Tool

• Web100 enhanced server handles testing and
  diagnostic services
• Java based and command line clients allows
  testing from any client (local or remote)
• Performance and configuration faults reported
  back to client
• Drill-down functions provide more details error
  reporting capabilities
• Grant from NIH/NLM to explore duplex mismatch
  detection

45
NDT Flow Diagram
46
Bulk Transport

• Build a library / tool for bulk transport that
  does not require kernel level modifications yet
  achieves the performance of such
• VFER library
• Congestion control hooks
• Implements loss-based congestion control
• Working on delay-based version
• File transfer utility
• An initial version demoed

47
Everything we work on is available

• Tools are open source, supported, well-documented
• BWCTL/Iperf, OWAMP, NDT are deployed across
  Abilene backbone and at many partners
• You can
• See ongoing measurement results at the Abilene
  Observatory
• Test to/from the Abilene backbone

48
Network Performance Measurement Workshops

• Example Course Materials
• http//e2epi.internet2.edu/npw/presentations.html
• Goals
• Grow installed base of BWCTL/Iperf, OWAMP, and
  NDT at GigaPoP and regional campuses.
• http//e2epi.internet2.edu/pipes/pmp/pmp-dir.html
• Begin integration into IT support processes.
• Create an installed base for perfSONAR
  deployment.
• Give each participant tool-specific cookbooks.

49
Network Performance MeasurementWorkshop
Locations and Dates

• Completed
• SOX / GaTech (03/05)
• CENIC / UCLA (06/05)
• JT Vancouver (07/05)
• OARNet / OSU (09/05)
• MAGPI / FMM (09/05)
• MAX / College Park (12/05)
• APAN (01/06)
• JT - Albuquerque (02/06)
• MERIT (02/06)
• Columbia / NYSERNet (04/06)
• University of Virginia (04/06)

• Planned
• Wisconsin (07/06)
• Under Consideration
• Alaska,