lambda station project andrey bobyshev; phil demar; matt crawford escc/internet2 winter 2008 joint...
TRANSCRIPT
Lambda Station
Project
Andrey Bobyshev; Phil DeMar; Matt Crawford
ESCC/Internet2 Winter 2008 Joint Techs
January 22; Honolulu, HI
http://www.lambdastation.org
Fermilab
Will
replace
as the most commonly purloined presentation prop?
recycled
Goals and major directions of the project
Software architecture, API, SOAP/XML, some details
How Lambda Station works - use cases
Status and the near-term plans
Outline of the talk
The goal of the project...
Target: last-mile problem between local computing facilities and alternate WAN paths
Strategy: develop a network path selection service to interface production storage & computing facilities with wide-area network paths. alternate network paths for high impact data movement selective forwarding on a per flow basis graceful cutover and fallback on path changes on-demand from applications (authentication & authorization) current implementation based on policy-based routing &
including the support of DSCP marking
LSInterfaceLS-Management &
Reporting Interface
mySQL:requests
, history,security
LSDIScovery
Service
LSRESource
Scheduler LSController
SiSi
local
definitions
online
updates
SiSi
SOAP/Clarens
Vendor specific
modules
NETWORK CONFIGURATOR
CISCO Force10 WAN
Storage & application spaceManagement
SOAP/Clarens
LSInterface
Remote Lambda Station
Data Exchange
Control & Management
Service-based Architecture: CLARENS is framework for service-
based architecture, mutual authentication
of requests LSController – synchronizes work of all
services, has control functions LSInterface - unified interface for
intercommunication between LS and
applications, and LS-to-LS LSDIScovery service – detects new
lambdastations, and PBR clients at remote
and local sites LSRESource Scheduler – estimates
bandwidth allocation, monitor real-time
usage of resources Network Configurator – dynamic
reconfiguring of LAN and WAN
SOAP
Software architecture
Check references at www.lambdastation.org Papers published at CHEP04, CHEP06, Gridnet2006,
CHEP07 Demonstrations at SC05, I2 FMM 2007; SC07
For more details...
Instead, we would like to walk you
through a few the real world use cases
Normal traffic flow
LambdaStation
@ Fermilab
Caltech Tier2
OSCARS
Internet2/ ESNet
Dynamic Circuits
WANLambdaStation
High Impact trafficControl messages
Network configuring
UNL Tier2
LambdaStation
US CMS Tier 1 Facility
UltraLight
Site Lambda StationSite Lambda Station
Network
infrastructureNetwork
infrastructure Routed R&E Networks
High Impact Data Network
Default network path
An alternate, high-impact path
Control plane
Use case 1: Alternate traffic
forwarding on-demand of Data
Movement Applications (dCache/SRM)
Two sites can exchange traffic via several networks. Each site has a Lambda Station to steer selected flows into an alternate
path on-demand of applications. Site networks can be dynamically reconfigured by Lambda Stations or statically pre-configured.
If supported by site network infrastructure or/and by applications QoS ToS or DSCP can be used to match selected flows.
CaltechFermilab
US CMS Tier1 Center Tier2 Facility
dCache/SRM
dCache/SRM
SRM@FNAL LS@FNAL NET@FNAL NET@CALTECH SRM@CALTECHLS@CALTECH
Configuring Network Configuring Network
High BW
path
Re-conf to default Re-conf to default
All time through the life cycle of reservation each site checks periodically the corresponding reservation at remote site. If
an error is detected, local reservation will be cancelled.
Data Movement
ReservationRequest BW
Local ID
ActiveOK
OK
OK
Status?
Request BW/path
Remote ID
ACK
Sequence Diagram - Use Case 1 (simplified)
Status?
Completed
Ticket expired
A path across DCN
Fermilab
Lambda Station Server
University of Nebraska
Network
infrastructureNetwork
infrastructure
Routed R&E Network
Default network path
Internet2 DCSESnet OSCARs
Lambda Station
Server
Control plane
Circuit call setup & teardownLAN reconfiguration to use circuitInter-Lambda Station coordinationFtWatch request/response
Use case 2: Setup alternate traffic path
between FNAL and UNL via Dynamic
Circuit based on flow analysis.
IDC/ESNet IDC/Internet2
Flow
data
Flow analysis
Sequence diagram - use case 2 (simplified)ftwatch LS@FNAL net@FNAL UNL LSIDC/I2net@UNL
OpenSvcTicket & call IDC
Remote ID
openSvcTicket
Status ?
OK
Reserve/Activate
OK
netconf
OK
netconf
OKPath is established
teardown
netconf netconf
Forward path across DCN
Site Lambda Station
Site A Site B
Network
infrastructureNetwork
infrastructure
General Internet
Default network path
Dynamic Circuits Networks
IDC-B
Reverse path across general Internet
IDC-A
Domain A Domain B
High Impact Data Network
Control plane
Circuit call setup & teardownLAN reconfiguration to use circuitApplication-initiated request/response
Use case 3: Sinking traffic via DCN or High Impact Networks
Functional system deployed at FNAL, UNL, & Caltech Application LS awareness - dCache/SRM Primitive protocol to request a DC via ESNet/Internet2 Successful demonstration SC05, I2 FMM, SC07 (Sort of…) Production use with US CMS Tier1, Tier2s
at UNL and Caltech Flow analysis to invoke a Lambda Station ticket based
on observed traffic
Summary – Current Status
Improve code to make production quality Support multi-platform clients Enhance protocol for interaction with OSCARs/ESnet &
DCN/I2 Improve distribution package Develop management/configuring tools Interface with monitoring system Documentation
Plans
Large-scale data recover via DCN
One week after I2 FMM demo, UNL lost their Tier-2 data cache
50TB Data recovered by transfer FNAL Tier-1: Largely via Internet2/ESNet
Dynamic Circuit Completed in 32 hours
Progress Chart of Lambda
Station Request in SC07
Demo
The END