Transcript 20071010-PerfUpdate

Performance Update
Jeff Boote
Senior Network Software Engineer
Internet2
Martin Swany
Assistant Professor
University of Delaware
The Distributed CI Computer
Instrumentation
Control
Help
Desk
Researcher
Security
Viewing
Data
Generation
Program
Security
Collab
Tools
Control
Training
Education
And
Outreach
Publishing
Policy and
Funding
Management
Security and
Access
Access
Control
Human
Support
Funding
Agencies
Authentication
Resource
Providers
Campuses
3D
Imaging
Display
Tools
Authorization
Security
Security
Analysis
Simulation
Computation
Program
Input
Archive
Retrieval
Data Sets
Storage
Search
Data
Directories
Schema
Metadata
Ontologies
Data
.
Input
Network
Security
Display and
Visualization
The Network is the Backplane for the
Distributed CI Computer
Instrumentation
Control
Help
Desk
Researcher
Security
Viewing
Data
Generation
Program
Security
Collab
Tools
Control
Training
Education
And
Outreach
Publishing
Policy and
Funding
Management
Security and
Access
Access
Control
Human
Support
Funding
Agencies
Authentication
Resource
Providers
Campuses
3D
Imaging
Display
Tools
Authorization
Security
Security
Analysis
Simulation
Computation
Program
Input
Archive
Retrieval
Data Sets
Storage
Search
Data
Directories
Schema
Metadata
Ontologies
Data
.
Input
Network
Security
Display and
Visualization
CI Components
Applications
Bulk
Transport
2-Way
Interactive
Video
Real-Time
Communications
….
….
….
Performance
Infrastructure / Tools
Middleware
Phoebus
Measurement
Nodes
Network
….
Control
Plane
Control Plane
Nodes
Network
Cyberinfrastructure
Applications call on Network Cyberinfrastructure
Internet2 Performance CI Software
• Performance Monitoring Infrastructure
• perfSONAR (with ESnet, GEANT2 JRA1, RNP,
many others)
• BWCTL, NDT, OWAMP, Thrulay
• Transport Infrastructure
• Phoebus
• Distributed System Infrastructure
• Topology Service (with University of Delaware)
• Distributed Lookup Service (with University of
Delaware, PSNC)
Overview
•
•
•
•
•
•
Performance Measurement Requirements
Measurement Tools
perfSONAR
perfSONAR-PS Services
Transport Middleware
Distributed System Infrastructure
• Schema for sharing
• Services for sharing (discovery/topology/authentication)
Performance Measurement Requirements (Internet2
network support)
•
Deployed on Internet2 Network
•
•
•
Meet the needs of the LHC Application Community (T0-T3)
•
•
Ease of deployment is key *
Veneer on existing NOC solutions rather than a new solution *
Integrated into a Cyberinfrastructure Solution
•
•
•
•
Partnership with GEANT2 JRA1, ESnet, RNP to meet that need
Packaged perfSONAR solution for US connectors and universities
•
•
•
Software: perfSONAR, BWCTL, NDT, OWAMP
Goals: IU NOC Support *, proof of performance for Application Community
Support DRAGON
Rely on common AA infrastructure
Common elements (AA, Discovery, Topology)
* Drives the need for a perl-based packaged solution, perfSONAR-PS
OWAMP (One way latency data)
•
3.0c (RFC 4645 version) available now
•
•
Maintenance mode
Diagnostic test points available at all Internet2 IP Network router
locations:
•
•
owamp.POP.net.internet2.edu
•
POP = losa,salt,hous,kans,chic,atla,newy,wash - soon seat
Full mesh of Internet2 IP network tests collected and available from:
http://owamp.net.internet2.edu/owamp_grid.cgi
BWCTL (Throughput tests)
•
1.2a is current version
•
•
•
2.0 due out soon with additional testers
Diagnostic test points available at all Internet2 IP Network router locations:
•
bwctl.POP.net.internet2.edu
•
POP = losa,salt,hous,kans,chic,atla,newy,wash - soon seat
Full mesh of Internet2 IP network tests collected and available soon
NDT updates
• Administrator can allow clients to run
repetitive tests
• MCNC modifications rolled into NDT code
• New post-processing tools to improve
analysis of log data
• New version available by end of this
week
Post-processing Tasks
• Allow site administrators to find problems
effecting multiple clients
• Allow site administrators to verify that
problems have been resolved
• Allow developers to experiment with new
detection algorithms
• Allow developers to quickly test new features
NDT Post-processing Tools
• Genplot
– Simply program to display Web100 data
• JAnalyze
– Java based version of Analyze program
Genplot – Textual Display
•
‘genplot -tmCurCwnd,SlowStart,CongAvoid,CongestionSignals fn | less’
Genplot – Graphical Display
•
‘genplot –C fn; xplot fn.xpl’
JAnalyze
NPToolKit
• Recent versions of Measurement tools
installed and pre-configured
• Knoppix Live-CD bootable system
• Current Version: 1.7
• http://e2epi.internet2.edu/networkperformance-toolkit.html
perfSONAR Motivation
• Most organizations can do monitoring and
diagnostics of their own network
• Networking is becoming an increasingly
cross-domain effort
• Monitoring and diagnostics must also become
a cross-domain effort
perfSONAR
• A set of protocols and schemas for implementing a
SOA for sharing and controlling network performance
tools
• A community of users and developers
• Multiple sets of interoperable software
perfSONAR-PS Motivation
• Create separate implementation of perfSONAR standard
• Use same protocol/standards
• Proof of interoperability (strengthens the standard)
• Targeted for University NOC deployments
• Lightweight
• Easy to deploy/manage
• (We were unable to convince our primary users to deploy Java
services due to the complexity of dependencies)
perfSONAR-PS Services
• Focus on development of major perfSONAR components
•
•
•
•
•
•
LS
Topology
Link Status
Circuit Status
SNMP Based MP/MA
PingER (SLAC)
• New additions
• OWAMP/BWCTL
• Traceroute
•
Contact Jason or Jeff if you are interested in testing pre-releases
Lookup Service
• 'Oracle' of perfSONAR deployments
• Accept service registrations
• Handle complex data/service queries
• Manage the lifetimes of data and services to keep
framework relevant
• Web Service interface to XML Database
• Berkeley DB XML
• Service Info/Data kept in native formats
• Draw away the complex query tasks from
otherwise 'busy' services
Lookup Service
• Also XML based configuration/protocol
• Native storage/query mechanisms [Xpath/XQuery]
• Message format to exchange the data
• Targeted at single domain deployment
• Single instance to manage multiple services
• Mutli-domain 'distributed' LS in development
• Client applications aware of LS to find services
• perfSONAR-UI
• perfAdmin
Lookup Service
• Current Deployment:
• Internet2 (Ann Arbor)
• University of Delaware
• Planned Deployment:
• IU for Internet2 network and regionals
• International Partners
Topology Service
• Provides a queryable repository for
obtaining topology information about a
domain
• Can obtain the entire network
• Xquery interface allows the construction of
arbitrarily complex queries about the network
• Topology is specified according to the
schema in development for OGF
Topology Service
• Current Deployments
• Internet2
• Planned Deployments
• Internet2 DCN
• SLAC (PingER Topology Information)
Link Status Measurement Archive
• Provide access to up/down status information
about layer2 links
• Data stored in a SQL database
• Database schema allows for storing time ranges during
which a link had a certain status
• Minimizes storage costs for rarely changing links
• Communication/Configuration via XML
• Target audience is network operators and users
interested in obtaining the status of the links over
which their data flows
Link Status Measurement Archive
• Collector
• Allows for the periodic collection of the status of
one or more links
• Can use SNMP, Scripts or simply Constants
• Can store results directly into a database or into
a remote Measurement Archive
Link Status Measurement Archive
• Visualization
• A perfSONAR-UI Plugin is available that can
display a network and the status of its links
• Current Deployment
• Internet2 Network
• HOPI (in2p3 circuit)
• Planned Deployment
• SLAC
Circuit Status Measurement Archive
• An e2emon-compatible service
• Integrates with the Link Status MA to provide the
information stored in MAs
• Can work with local MAs directly or with remote MAs
• Can use the Topology MA to obtain necessary information
about nodes
• Can use a Lookup Service to lookup the MA containing
information on each link
• Target audience is administrators who want to
publish circuit status information to e2emon clients
Circuit Status Measurement Archive
• Visualization
• Any tool that is compatible with e2emon will
work with this service
• Current Deployment
• Internet2 Network
• HOPI (in2p3 circuit)
• Planned Deployment
• SLAC
SNMP Measurement Archive
• Provide access to network performance data
• Utilization
• Errors
• Discards
• Numerous tools exist to collect passive
measurements (via SNMP):
• MRTG
• Cacti
• Cricket
• Expose archives from RRD files
SNMP Measurement Archive
• XML based configuration/protocol
• Configuration file to expose the data
• Message format to exchange the data
• Targeted at researchers, NOC staff debugging
intra/inter domain network problems
• Many client applications available to graphically
display the results:
• perfSONAR-UI
• perfAdmin
• perfOMeter (General Session Demo)
SNMP Measurement Archive
• Current Deployment:
•
•
•
•
Internet2 Network
ESnet
Georgia Tech/SOX
Fermilab
• Planned Deployment:
• Regional Networks
• International Partners
Visualization
• Utilizing the plugin architecture of perfSONAR-UI
• Data visualization beyond network utilization
• Google Maps
• Utilization by physical location
• 'Weather Map' of Internet2 Network
• Web based speedometer to interact directly with MA
code
Other Services (in development)
• Traceroute MP/MA
• Joint effort between UD and SLAC
• Integrate into visual traceroute tools
• Currently in testing
• OWAMP MA
• Make regular results available
• BWCTL MA
• Make regular results available
perfSONAR-PS Releases
• Currently exploring testing and packaging
strategies
• Releases will ultimately be CPAN based
• Standing issues include configuration/automation
help.
• Versions of SNMP MA and LS will be
available this fall
• Pre-CPAN versions will be available for LHC
commitments in August (using CPAN
conventions)
Transport Middleware (Phoebus)
• The Phoebus project aims to help bridge the performance gap
by bringing dynamic networks to end-users
• Phoebus is another name for the mythical Apollo in his role as the
“sun god”
• Phoebus is based on the concept of a “session” that enables
multiple adaptation points in the network
• Improves end-to-end throughput for users
• Depots on the ingress/egress points of backbone networks
mitigate the effects of edge networks
• Allows transparent use of network features like cross-domain
dynamic circuits
Session Layer
• A session is the end-to-end composition of segmentspecific transports and signaling
• More responsive control loop via reduction of signaling
latency
• Adapt to local conditions with greater specificity
• Buffering in the network means retransmissions need not
come from the source
Session
Transport
Network
Data Link
Physical
User Space
Transport
Network
Data Link
Physical
Session
Transport
Network
Data Link
Physical
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
End-to-End
Session
Optical Net

QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
End-to-End
Session
Optical Net
Phoebus - Upcoming SC Demo
• Allocate dynamic circuits using DCN
Control Plane
• UDT for inter-depot communication
• Transparent operation (library, iptables)
• Legacy applications can use the DCN
with no modification
Phoebus - Current Work
• Separating front-end (control plane)
from back-end (data plane)
• Allows creation of new front-end services
• Constructing a web services based
front-end
• Andy Lake for Google Summer of Code
• Construction of high-performance
forwarding devices with myricom NICs
Phoebus - Future
• Simple file transfer tool
• Transparently use Phoebus/Dynamic
Circuits
• Utilize Measurement Infrastructure
• Help find best routes, provide information
about paths and achievable bandwidth
• Extension of Path Finding / Routing
• Authentication and Authorization
Schema/Protocol Developments
• The perfSONAR Topology schema is also used in the
DCN control plane
• We’ve spent quite a bit of effort harmonizing these
• The obvious win is that we have the measurement
system have immediate access to dynamic circuits
• The broader impact is that we’re approaching a
unified network interaction model (UNIM)
Schema Developments
• The perfSONAR schema is being
standardized in the Open Grid Forum
(OGF) Network Measurement Working
Group (NM-WG)
• The topology efforts have spawned a
new WG called the Network Markup
Language WG (NML-WG)
Schema - Network Element Identifiers
• A scheme for identifying network
elements
• Each network element gets a unique
identifier
• This identifier will be included with any
measurement associated with that
element.
Network Element Identifiers
• Use Cases:
• A topology service can be used to find the
identifier for a network element
• An LS could then be queried to find all
measurements associated with that element
• Dynamic service path-finding can be
integrated with ongoing measurements
Network Element Identifiers
• Identifiers use URN notation
• Prefixed with “urn:ogf:network:”
• Consists of name/value pairs separated by
colons
• Possible field names: domain, node, port,
link, path, network
• Set of rules defined for each field to keep
identifiers compact and finite
Network Element Identifiers
• Examples
•
•
•
•
•
•
•
urn:ogf:network:domain=Internet2.edu
urn:ogf:network:domain=internet2.edu:node=packrat
urn:ogf:network:domain=internet2.edu:node=rtr.seat:port=so-2%2F1%2F0.16
urn:ogf:network:domain=internet2.edu:node=rtr.seat:port=198.32.8.200
urn:ogf:network:domain=Internet2.edu:node=packrat:port=eth0:link=1
urn:ogf:network:domain=internet2.edu:link=WASH to ATLA OC192
urn:ogf:network:path=anna-11537-176
Distributed Systems Infrastructure
• perfSONAR, DCN and Phoebus have similar
system requirements
• Lookup and Topology Services comprise a generic
Information Service that is useful to all these
Network Services
• Authentication and Policy services are crosscutting as well
• Rather than have silos of mission-specific
functionality, we envision pervasive system
components
Distributed Systems Infrastructure
• Synergies of information bases are
obvious
• Multi-layer path-finding including current
network state, available resources on a
variety of layers
• It is a compelling vision to imagine a
dynamic, reactive, visible service-rich
network
Summary
• A rich set of tools are being developed
• To federate network monitoring and diagnostics
• To enable dynamic network resource allocations
• To leverage new network capabilities from an ‘end-user’
application (phoebus)
• A longer view toward an evolution of “in the network”
services
Questions?
• Jeff Boote
• [email protected]
• Martin Swany
• [email protected]