Longer version

Grids for Real-time and

Streaming Applications

GCC2005 Beijing China

December 1 2005

Geoffrey Fox

Computer Science, Informatics, Physics

Pervasive Technology Laboratories

Indiana University Bloomington IN 47401

http://gr

ids.ucs.indiana.edu/ptliupages/presentations/

Four Data Streaming Application Areas

n

Data Assimilation

applied to link the

data deluge

(satellites, sensors, seismometers) in real time to small

and large scale parallel simulations

•

Use in Earthquake Science

n

Department of Defense

(and

Homeland Security

) have

built the

Global Information Grid

with a target

architecture

NCOW

(Network Centric Operations and

warfare)

•

They

submit no jobs

; rather stream data to brokers from

which they are filtered and distributed

•

Includes their rather dated

distributed simulation HLA

n

Audio-Video Conferencing

implemented with services

and Grid messaging

n

Hand-held Grid

linking PDA/cell-phones to Grids

Database

S S

S

S SS SS SS SS SS SS SS SS

F S F S F S F S F S F S F S F

S SF F S F S F S F S F S F S F S F S F S F S F

S

Por

tal

MetaData Filter Service Sensor Service Other Service Anothe Grid

Raw Data  Data  Information  Knowledge  Wisdom

Decisions S S S S Anothe Service Anothe Service S S Anothe

Grid S S

Key Concepts

n

Grid of Grids

(System of Systems) allows “library”

approach to composing Grids

n

Service Oriented architectures

(Web or Grid services)

are attractive for many/most distributed systems

n

There are many applications that are

NOT

best

considered as

jobs and files

(classic Grid) but rather as

streams and filters

(services)

n

Services exchanging messages

become

Services exchanging streams

(sets of messages)

n

Publish-Subscribe

messaging gives better QoS and

management than point to point messaging with

negligible performance loss

n

Always use standards including those for

GIS

Typical use of Grid Messaging in NASA

Datamining

Grid

Sensor Grid

Grid Eventing

GIS Grid

GIS and Sensor Grids

n

OGC

has defined a suite of

data structures

and

services

to support

Geographical Information Systems and

Sensors

n

GML

Geography Markup language defines

specification of geo-referenced data

n

SensorML

and

O&M

(Observation and Measurements)

define meta-data and data structure for sensors

n

Services like

Web Map Service, Web Feature Service,

Sensor Collection Service

define services interfaces to

access GIS and sensor information

n

Grid workflow

links services that are designed to

support streaming input and output messages

n

We are building Grid (Web) service implementations of

these specifications for NASA’s

SERVOGrid

Grid Eventing via NaradaBrokering

n

The

Scripps Orbit and Permanent Array Center

Grid Messaging with Grid of Grids Architecture

HPSearc

h

Manages

Narad

Brokering

Sensor Grid

WS-Context

Stores dynamic

data

WFS

(GIS data)

Post befor

Processing

Post afte

Processing

Notify

Subscribe

Databas

e

Archives

Web Feature Service

GIS Grid

DoD Principles

Filter or

Datamining Grid

WMS uses WFS that uses data sources

<gml:featureMember>

<fault>

<name>

Northridge2

</name>

<segment>

Northridge2

</segment>

<author>

Wald D. J.

</author>

<gml:lineStringProperty>

<gml:LineString

srsName="null">

<gml:coordinates>

118.72,34.243

-118.591,34.176

</gml:coordinates>

</gml:LineString>

</gml:lineStringProperty>

</fault>

</gml:featureMember>

Google Map Client

Google Central

Google

Map

Client

UDD

I

WFS

2

Databases wit

SERVOGrid Faults

WFS

1

SOAP

Sensor

Grid

HTTP

Helper

Service

s

Archive

d

Real Time

Real Time GPS

and Google Maps

Subscribe to live GPS

station. Position data

from SOPAC is

combined with Google

map clients.

Select and zoom to

GPS station location,

click icons for more

information.

Google maps

can be

integrated with

Web Feature

Service

Archives to

filter and

browse seismic

records.

Integrating

Archived Web

Feature Services

Google Maps

as Service

accessed from

our WMS

Google plus GIS Grid Integrated with Los Alamos

Critical Infrastructure Simulations for DHS

Natural Gas Layer

Energy Power Layer

NaradaBrokering 2003-2006

n

Messaging infrastructure

for

collaboration, peer-to-peer

and

Grids

Implements

JMS

and

native

high-performance

protocols (

message

transit

time of 1 to 2 ms per hop

)

n

Order-preserving

message

transport

with

QoS

and

security

profiles

Support for different underlying

transport

such as

TCP, UDP,

Multicast, RTP

n

SOAP message support

and

WS-Eventing, WS-RM

and

WS-Reliability

.

•

WS-Notification

when specification agreed

n

Active replay

support: Pause and Replay live streams.

n

Stream Linkage:

can link permanently multiple streams – using in

annotation of real-time video streams

n

Replicated storage support

for fault tolerance and resiliency to storage

failures.

n

Management:

HPSearch Scripting Interface to streams and brokers

(uses

WS-Management

)

n

Broker Topics and Message Discovery:

Locate appropriate

Integration with

Axis2 Web Service Container

(?)

n

High Performance Transport

supporting SOAP

Infoset

0

1

2

3

4

5

6

7

8

9

100

1000

Transit

Delay

(Milliseconds)

Message Payload Size (Bytes)

Mean transit delay for message samples in

NaradaBrokering: Different communication hops

hop-2

5

hop-7

hop-3

Pentium-3, 1GHz,

256 MB RAM

100 Mbps LAN

Average Video Delays for one broker –

Performance scales proportional to number of brokers

Latency ms

# Receivers One session Multipl

sessions

30 frames/sec

Consequences of Rule of the Millisecond

Useful to remember

critical time scales

•

1)

0.000001 ms

– CPU does a calculation

•

2a)

0.001 to 0.01 ms

– Parallel Computing MPI latency

2b)

0.001 to 0.01 ms

– Overhead of a Method Call

•

3)

1 ms

– wake-up a thread or process (do simple things

on a PC)

•

4)

10 to 1000 ms

– Internet delay

n

2a), 4) implies geographically distributed

metacomputing

can’t in general compete with parallel

systems

n

3) << 4) implies a software overlay network is possible

without significant overhead

•

We need to explain why it adds value of course!

n

2b) versus 3) and 4) describes regions where

method

and

message

based programming paradigms important

HPC Simulation

Data

Filter

Data

Filter

Data

Filt

er

Data

Filt

er

Data

Filt

_er

Distributed Filters

massage data

For simulation

Oth

er

Gri

an

d W

eb

Servi

ces

Analysi

Control

Visualize

SERVOGrid (Complexity) Computing

Model

Grid

OGSA-DA Grid Services

This Type of Grid

integrates with

Parallel computing

Multiple HPC

facilities but only

use one at a time

Many simultaneous

The Ten areas covered by the 60 core WS-* Specifications

WSRP (Remote Portlets)

10: Portals and User Interfaces

WS-Policy, WS-Agreement

9: Policy and Agreements

WSDM, WS-Management, WS-Transfer

8: Management

WSRF, WS-MetadataExchange, WS-Context

7: System Metadata and State

UDDI, WS-Discovery

6: Service Discovery

WS-Security, WS-Trust, WS-Federation, SAML, WS-SecureConversation

5: Security

BPEL, WS-Choreography, WS-Coordination

4: Workflow and Transactions

WS-Notification, WS-Eventing (Publish-Subscribe)

3: Notification

WS-Addressing, WS-MessageDelivery; Reliable Messaging WSRM; Efficient Messaging MOTM

2: Service Internet

XML, WSDL, SOAP

1: Core Service Model

Examples

WS-* Specification Area

Activities in Global Grid Forum Working Groups

Authorization, P2P and Firewall Issues, Trusted Computing

7: Security

Resource/Service configuration, deployment and lifetime, Usage records and access, Grid economy model

6: Management

Network measurements, Role of IPv6 and high performance networking, Data transport

5: Infrastructure

Database and File Grid access, Grid FTP, Storage Management, Data replication, Binary data specification and interface, High-level publish/subscribe, Transaction management

4: Data

Job Submission, Basic Execution Services, Service Level Agreements for Resource use and reservation, Distributed Scheduling

3: Compute

Software Interfaces to Grid, Grid Remote Procedure Call, Checkpointing and Recovery, Interoperability to Job Submittal services, Information Retrieval,

2: Applications

High Level Resource/Service Naming (level 2 of fig. 1), Integrated Grid Architecture

1: Architecture

GS-* and OGSA Standards Activities

GGF Area

The Global Information Grid Core Enterprise Services

Provisioning, operations and maintenance of applications.

CES9: Application

Retention, organization and disposition of all forms of data

CES8: Storage

Includes automated and manual methods of optimizing the user GiG experience (user agent)

CES7: User Assistance

Provision and control of sharing with emphasis on synchronous real-time services

CES6: Collaboration

Includes translation, aggregation, integration, correlation, fusion, brokering publication, and other transformations for services and data. Possibly agents

CES5: Mediation

Searching data and services

CES4: Discovery

Synchronous or asynchronous cases

CES3: Messaging

Supports confidentiality, integrity and availability. Implies reliability and autonomic features

CES2: Information

Assurance (IA)/Security

including life-cycle management

CES1: Enterprise Services Management (ESM)

Service Functionality

Core Enterprise Services

DoD Core Services and WS-* plus GS-* I

NCOW Capability Interfaces, JSR168 GridSphere

WS-10 Portlets

CES 7: User assistance

XGSP, Shared Web Service ports, Anabas Asynchronous Virtual

Organizations Shared Web Resources

CES 6: Collaboration

Treatment of Legacy systems. Data Transformations

WS-4 Workflow

CES 5: Mediation

Extended UDDI WS-6 UDDI

CES 4: Discovery

NaradaBrokering, Streaming/Sensor Technologies WS-2, WS-3 Service Internet Notification

CES 3: Messaging

Grid-Shib, Permis Liberty Alliance etc.

GS-7 Securit (Authorization) WS-5

WS-Security

CES 2: Information Assurance(IA)/Security

CIM GS-6: Management

WS-8 Management

CES 1: Enterprise Services Management

B: NCOW Core Services (to be continued)

Legacy subsystems and modular architecture

Grid of Grids Composition

Industry Best Practice (IBM, Microsoft …)

Build Grids on Web Services

WS-1: Core Service Model

Use Service Oriented Architecture

A: General Principles

Others GGF

WS-* Service area NCOW Service or Feature

DoD Core Services and WS-* and GS-* II

OGC GIS standards

Geographical Information Systems GIS

OGC Sensor standards Work starting

Sensors (real-time data)

Extend computer scheduling to networks and data flow

Distributed Scheduling and SLA’s (GS-3)

Resource/Service Matching/Scheduling C2IEDM, XBML, DDMS, WFS Semantic Grid Globus MDS WS-7 System Meta-data

C: Key NCOW Capabilities not directly in CES

WS-9 Policy

Environmental Control Services ECS

Best Practice in building Grid/Web services (proxy or direct)

GS-2; invoke GS-3

CES 9: Application

NCOW Data Strategy GS-4 Data

CES 8: Storage (not real-time streams)

B: NCOW Core Services Continued

Others GGF

WS-* Service area NCOW Service or Feature

GlobalMMCS SWT Client

Chat

TV

Webcam

Video

Mixer

GIS

e - Annotation Playe

r

Archived stream playe

r

Annotatio

nplaye / WB r

Archieved stream

list

Real time stream

list

e -Annotation Whiteboar

d

Real time stream playe

r

Archived

Real Time

Real Tim

Stream List

Stream List

Player

e-Annotation

Archived Stream

Annotated

e-Annotation

Player

Player

Stream Player

Whiteboard

PDA Download video (using

4-way video mixer service)

PDA

Desktop

Location of software for Grid Projects in

Community Grids Laboratory

n

htpp://www.naradabrokering.org p

rovides Web service

(and JMS) compliant

distributed publish-subscribe

messaging

(software overlay network)

n

h

tpp://www.globlmmcs.org is

a

service oriented (Grid)

collaboration environment

(audio-video conferencing)

n

ht

tp://www.crisisgrid.org is

an OGC (open geospatial

consortium) Geographical Information System (GIS)

compliant

GIS and Sensor Grid

(with POLIS center)

n

htt

p://www.opengrids.org has

WS-Context, Extended

UDDI etc.

n

The work is still in progress but NaradaBrokering is

quite mature