overview

G. Maron, EuChinaGrid 1st Conference, Rome, September 18-19, 2006

Grid enabled remote instrumentation and

sensors with distributed control

The GridCC Project

Gaetano Maron

Istituto Nazionale di Fisica Nucleare

Laboratori Nazionali di Legnaro, Legnaro Italy


Overview

• Bringing Instrument into the Grid: the Instrument Element

• The GridCC Project: Introduction

• The GridCC Test-bed: Pilot applications


The Grid Technologies to extend the limit of a single computer (center)

User Interface

ComputingElement

ComputingElement

ComputingElement

StorageElement


Extending the Grid Concepts

Satellite viewsto monitor the volcano

Control and Monitor RoomTo model calculations

and disaster predictions

Terrestrial probes to monitorThe volcano activities

Instrument Element


The GridCC Project

Instruments Grid

Computational Grid

Data for Model Calculations

Predictions

Control and Monitor Room


General on the GridCC ProjectParticipant name Country

Istituto Nazionale di Fisica Nucleare Italy

Institute Of Accelerating Systems and Applications

Greece

Brunel University UK

Consorzio Interuniversitario per Telecomunicazioni

Italy

Sincrotrone Trieste S.C.P.A Italy

IBM (Haifa Research Lab) Israel

Imperial College of Science, Technology & Medicine

UK

Istituto di Metodologie per l’Analisi ambientale – Consiglio Nazionale

delle Ricerche

Italy

Universita degli Studi di Udine Italy

Greek Research and Technology Network S.A.

Greece

• It is a 3 years project. Started the 1st September 04

• Funded by EU in the Frame Program 6

• 10 Partners from 3 EU Countries + (Israel)

• About 40 people engagged

• www.gridcc.org


StorageElementsStorage

Elements

ComputingElement

ComputingElement

InstrumentElement

Instrument Element: global scenario

ComputingElement

StorageElement

InstrumentElement

InstrumentElement

Existing Grid Infrastructures

Web ServiceInterface

Virtual Control Room


Exec. Service

WfMS

WMS

AgrSUser direct

ActionIndirect Action

SE



(VCR)

All end user access is via the

VCR

Instrument elements

(IE)

The IE is a virtualization of the real physical instrument

Instrument elements

(IE)Instrument elements

(IE)

Of course there may be many IEs

Compute and Storage Elements (with advanced reservation)

StorageElement

(SE)

Compute element

(CE)

Of course Many CEs and SEs

StorageElement

(SE)

Compute element

(CE)StorageElement

(SE)

Compute element

(CE)

CollaborativeServices

(CS)


(VCR)

Users generally not working alone

Direct access to IE

SE (and CE) possible but often not desirable

Information and Monitoring

Services(IMS)

“Fast” all pervasive messaging system

Information System

(IS)Slowly updating information

Security Services

Security is essential to the success of the project

Global ProblemSolver

Watching (via the IMS) for problems anywhere in the system and acting to resolve them.

Execution Services

More complex workflows, including advanced reservation and QoS guarantees , allowed

The GridCC Architecture


IE RequirementsWeb Services

Instrument Element

Any Protocol or physical connection

Sensor Network

Instrument

Instrument

Grid

ComputingComputing ElementElement

StorageStorage ElementElement

ComputingComputing ElementElement InstrumentInstrument

ElementElement

1: Provide a uniform access to the physical device

2: Allow a standard grid access to the instruments

3: Allow the cooperation between different instruments that belong to different VOs


Instrument Element: a Black Box

IEVIG

SCommands

Status

Fast communication channel

• The term Instrument Element describes a set of services that provide the neededinterface and implementation that enables the remote control and monitoring of physical instruments.

Grid Interaction

SR

M/S

E

Instruments

Quick Answers to the previous slide: 1) The VIGS provide a Web Service acccess to

the instrumentation2) The fast communication channel disseminate

the acquired information between instruments3) The Data Mover provide a SRM/SE interface


Device Virtualization Model

Instrument

Parameters

Attributes

Control

Model

XML Based

Language

1. Parameters hold configuration information 2. Attributes hold instrument variables 3. Control Model hold actions 4. XML Based Language to allow the device to describe itself

• Parameters: Maximum Voltage, Minimum voltage• Attributes: measured Voltage• Commands: Perform a measure

Voltmeter


Instrument Instrumentation

getContextsgetInstrumentManagersgetInfo

getIstanceget/Set ParametersgetCommandsexecuteCommandgetStategetStateMachine

IEV

IGS

lockInstrumentsunlokInstrumentsretrieveLoked

getRemoteExecutionTimegetOneWayCostgetTotalMethodExecutionTime

Instruments

We can divide the Instrumentation in 3 main parts: • The direct access to the Instruments• The advance instrument reservation (interaction with the Agreement Service (AS)) in order to achieve (hard) guarantees• The Possibility to predict the execution time of the instrumentation methods in a concurrent access (soft guarantees)

Instrumentation method Documentation http://sadgw.lnl.infn.it:2002/IEFacade

Crucial non-Functional Requirements: • Instruments could be order of 106

• Only authorized people should access to the instruments of a VO• The instrumentation is not a batch process like a job submission! Interactivity is mandatory

• A Distribute and hierarchic implementation is mandatory • the Security overhead should be negligible

http://sadgw.lnl.infn.it:2002/IEFacade


Instrument Element ArchitectureVirtual Instrument Grid Service (VIGS)

ResourceService

Inf & MonService

ProblemSolver

InstrumentManager

Instrument Element

Data Mover

IMSProxy

ControlManager

DataCollector

Real Instruments

Data Flow

Control Flow

State FlowError FlowMonitor Flow

• The term Instrument Element describes a set of services that provide the needed interface and implementation that enables the remote control and monitoring of physical instruments.

Acc

ess

Con

trol M

anag

er

execute()

getState()

create()

destroy()

InputManager

EventProcessor

FSMEngine

ResourceProxy

Control Manager


Interacting with Instrument Elements1) Full GridCC Environment

2) Partial GridCC Environment

3) Standalone Environment

IMS

ComputingElementComputingElement

ComputingElement

StorageElements

StorageElements

StorageElementGPS

VCR

Instrument

Element

Instrument

Element

Instrument

Element

Security

Service

Exec

utio

nSe

rvice

IMS

ComputingElement

ComputingElement

ComputingElement

StorageElements

StorageElements

StorageElementGPS

Instrument

Element

Instrument

Element

Instrument

Element

VCR

Security

Service

Exec

utio

nSe

rvice

IE is web service based, any web service compliantclients can reach it. This mode of work is very usefull for small systems and for prototyping and debug large systems

Instrument

Element

Instrument

Element

Instrument

Element

VCR

Secur ity

Ser vic e

. NET

Visu

al Ba

sicVi

sual

C++

WS - I

Perl

C ++

Java

JSP

This mode of operation can be used when the application does not need to access CEs and SEs. It coud for instance exploit the workflow manager of the execution service to do unattended cycles of operations and control the system via VCR


Instrument Element Implementations

ResourceService

Inf & MonService

ProblemSolver

InstrumentManager

Instrument Element

Data Mover

Acc

ess

Con

trol M

anag

er

The IE components are typically implemented into a fully equipped Machines (e.g. dual core cpus, large memory, large disks, etc). This is true for RS, IMS and PS. For IM (and DM) there are 2 possibilities, according to the application type:• IM implemented in a fully equipped machine• IM embedded into the instrument that should be controlled

IM

RSIMS

IM

IM

IM

Embedded Web Service


Instrument Element on a Chip


Instrument Manager

IM is composed by 3 main components:- Control Manager:

- Input Manager. It handles all the input events of the IM. These includes commands from GUIs or other IMs, errors/state/log/monitor messages. - Event Processor. It handles all the incoming message and decide where to send them. It has processing capability - FSM. A finite state machine is implemented - Resource Proxy. It handles all the outgoing connections with the resources.

- Data Collector. It get data from the controlled instruments and make them available to the data mover. A local storage of the data is even foreseen.

- IMS Proxy. It receives error/state/log/monitor information from the controlled resources and forward them to IMS

IMSProxy

DataCollector

Instrument Manager

InputManager

EventProcessor

FSMEngine

ResourceProxy

Control Manager

Instruments

Data FlowState Flow

Error Flow

Monitor Flow

Control Flow

Customizable Plug-in modules to interface to the instruments


Resource Service Architecture

• The Resource Service (RS) handles all the resources of an IE and manages their partition (if any). • A resource can be any hardware or software component involved in the IE (instruments, Instrument

Managers, IMS components)• RS stores the configuration data of the resources and download them to resource target when necessary• Resources can be discovered, allocated and queried. • It is the responsibility of the RS to check resource availability and contention with other active partitions

when a resource is allocated for use. • A periodic scan of the registered resources keeps the configuration database up to date.• RS is interfaced to the WMS

DiscoveryManager

SubscribeManager

Partition&LockManager

ConfigurationManager

Available Resources

PartitionDefinitions

ConfigurationDefinitions R

S D

ata

Bas

es

Partition/Configurationretrieve methods

Partition and Locksetting methods

Configurationsetting methods

Discoverymethods


Information and Monitor System (IMS)

PUBLISHERS(Instruments nodes)SUBSCRIBERS

Errors Log infoMonitorState

• The Information and Monitor Service (IMS) collects messages and monitor data coming from GRID resources and supporting services and stores them in a database. There are several types of messages collected from the sub-systems. The messages are catalogued according to their type, severity level and timestamp. Data can be provided in numeric formats, histograms, tables and other forms.

• The IMS collects and organizes the incoming information in a database and publishes it to subscribers. These subscribers can register for specific messages categorized by a number of selection criteria, such as timestamp, information source and severity level.

Instrument

Manager

Inst

rum

ents

Instrument

Manager

Inst

rum

ents

Instrument

Manager

Inst

rum

ents


Problem Solver

IMSProxy

ControlManager

Instrument Manager

Pub/Sub

IMSProxy

ControlManager

Instrument Manager

IMSProxy

ControlManager

Instrument Manager

IMSProxy

ControlManager

Instrument Manager

DBData Mining Tools

Algorithms evaluations :Rule Induction, Tree, Functions, Lazy, Clusters and Associative

State FlowError FlowMonitor Flow

On Line Analisys

Problem Solver

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

iris

glas

s

brea

st c

ance

r

bupa

votin

g-re

cord

s

hous

ing

bala

nce-

scal

e

Bre

ast

Can

cer W

isco

nsin

Pim

a-In

dian

s-D

iabe

tes

tic-ta

c-to

e

Seg

men

t

Seg

men

tatio

n

Sic

k-eu

thyr

oid

Pag

e-B

lock

s

mus

hroo

m

Shu

ttle(

2)

Lette

rRec

ogni

tion

krko

pt

Shu

ttle(

1)

conn

ect-4

dataset

accu

racy

Average Rule AccuracyAverage Tree AccuracyAverage Function AccuracyAverage Instance AccuracyAverage Cluster Accuracy

Step 1 The control manager can perform an autonomous recovery action where the cost for the determination it is not so heavy .

Step 2 Persistent information can be analyzed in order to extract knowledge

Step 3 On-line information can be analyzed in order to detect possible malfunctions


Fast Instrument Communication Channel


Message Oriented Middleware Topic A

Topic B

• Subscribers Subscribe to a given Topic with a subscribe condition

• Publisher publish message in asynchronous way with a given message condition

• Publisher and subscribers can be part of the same program or in WAN distributed machines

• JMS Provide a standard set of API that standardize this communication system

• Many Commercial and academic implementation of this API exist in both C/C++ and Java (NaradaBrokering, Sun, IBM, SonicMQ etc etc )

In Our Case: • Each instrument can be a data

publisher or a data consumer • For more demanding application

an instrument must send/receive data in a streaming way


RMM-JMS• RMM-JMS is a JMS implementation on top of our high performance Reliable Multicast

Messaging (RMM) layer which provides one-to-one, one-to-many data delivery or many-to-many data exchange, in a message-oriented middleware point-to-point or publish/subscribe fashion

• The exceptional performance supports remote and distributed control and operation of scientific instruments such as sensors and probes

• Multicast transport for publish/subscribe messaging: Supporting the JMS Topic-based messaging and API, with matching done at the IP multicast level. The transport is a Nack-based reliable multicast protocol.

• Direct (broker- less) unicast for point-to-point messaging: JMS Queues are implemented over RMM queues. The transport is the TCP protocol.

• Brokered unicast transport for publish/subscribe messaging. The broker receives messages from the producer in either unicast or multicast delivery mode, and sends the messages to the subscribers in either mode

• broker serves as a bridge in a LAN-WAN-LAN configuration

Main Contribution of IBM Haifa Research Lab (Israel)


Performance: message rate – the many-to-one

• Blade center with 12 CPUs and 1GB Ethernet switch • No message loss • Total throughput: 61MBytes/sec. and 67MBytes/sec. for (a) and (b) respectively

(b) rate - msg size 100000 bytes

0

100

200

300

400

500

600

700

800

900

1000

0 5 10 15 20Number of Publishers

msg

/sec min

MaxAvgSDev

(a) rate - msg size 1000 bytes

0

10000

2000030000

4000050000

60000

70000

80000

90000

100000

0 5 10 15 20

Number of Publishers

msg

/sec min

Max

Avg

SDev


Performance: message rate – the one-to-many

• Blade center with 12 CPUs and 1GB Ethernet switch • No message loss • Peak result of over than 400000 msg/sec. was reached

Rate, msg size 1 Byte

0

100000

200000

300000

400000

500000

600000

0 5 10 15 20 25 30Number of Subscribers

msg

/sec min

Max

Avg

SDev

Rate, msg size 1000 bytes

0

10000

2000030000

40000

50000

6000070000

80000

90000

0 5 10 15 20 25 30

Number of Subscribers

msg

/sec min

MaxAvgSDev


Performance: round trip time (RTT, Latency)

• Two machines with a single publisher and a single subscriber on each one • Average round trip time computed over 1000 samples

RTT

0.01

0.1

1

10

100

1 10 100 1000 10000 100000 1000000Messages Size

Tim

e (m

Sec)

AvgSdevPing


Standard Grid Interaction


Data Mover

• The task of this element is to get data from the “data collector” of the IM• Data can be accessed via:

– Web service interface for generic data dump (e.g. slow storage, spy stream, etc.)– grid storage element (SE) and available CEs can access to the data via an SRM

Interface– Http server and TCP communication for high performance had-hoc data transfer

• The Data Mover exposes its methods to the IE web service and can be instrumented itself as an instrument.

Instrument Resources

DataMover

DataCollector

IM

IE Web Service Interface: get_data()

SRM interface

Http Server andTCP/IP raw socketData

Collector

IM

DataCollector

IM


Current IE Implementation a fist taste


Instrument Manager Performances (I)Instrument Manager Invocations

0

10

20

30

40

50

60

1

HTTP Transport Layer

Invo

catio

n pe

r S

econ

d

Average

min

Max

Variance

Asyncronous msg Rate

0

50

100

150

200

250

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Number of Client

Mes

sage

s pe

r S

econ

d

Average

min

Max

Variance

Virtual Instrument Grid Service (VIGS)A

cces

s C

ontro

l Man

ager

execute()

getState()

create()

destroy()

IMSProxy

DataCollector

InputManager

EventProcessor

FSMEngine

ResourceProxy

Control Manager

Test 1

Test 2

Test 1: Web Service invocation and status switch of FSM

Test 2: Soap Server receiving XML message format. DOM based parser

Virtual Instrument Grid Service (VIGS)A

cces

s C

ontro

l Man

ager

Acc

ess

Con

trol M

anag

er

execute()

getState()

create()

destroy()

IMSProxy

DataCollector

InputManager

EventProcessor

FSMEngine

ResourceProxy

Control ManagerIMSProxy

DataCollector

InputManager

EventProcessor

FSMEngine

ResourceProxy

Control Manager

Test 1

Test 2

Test 1: Web Service invocation and status switch of FSM

Test 2: Soap Server receiving XML message format. DOM based parser


Instrument Manager Performances (II)Command Distribution Time

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

10 50 80 120

Number of Instrument

Ave

rage

Tim

e (s

ec) 1FM

1FM M

3FM

3FM M3FM 3PC

3FM 3PC M

1

2

3

1 + 2

3

1

3

1

Optimized environment

IM with CMS Instruments


Summary Test output disabled

0

1000

2000

3000

4000

5000

6000

7000

8000

0 2 4 6 8 10 12 14 16 18 20 22

n publisher clientsm

sg/s

ec

xdaq

java

c++ (not xdaq)

java-C++contemporary

IMS

Errors/log/states messages(xml and java objs)

DB

TCP/IPPub/Sub

(JMS)

WebService

Interface

Summary Java vs Xdaq( C++) 1 socket output

0500

1000

15002000

25003000

35004000

0 2 4 6 8 10 12 14 16 18 20 22

n publisher clients

msg

/sec java

xdaq

JMS increasing the subscribers number

0

100

200

300

400

500

600

700

0 2 4 6 8 10 12

n subscriber clients

msg

/sec

w ith selectorfield

w ithoutselector field

Summary Mysql vs Oracle

0

500

1000

1500

2000

2500

0 2 4 6 8 10 12 14 16 18 20 22

n publisher clients

msg

/sec

Oracle 100Oracle 1

Mysql 100 Mysql 1

IMS Performances

IMSProxy IMS

Proxy

IMSProxy

….


Main IE Pilot Applications: Power Grid

Instrument Manager

Instrument Element

...



Gas

Solar

Power Grid V.O


Main GridCC Pilot Applications: Control and Monitor of high energy experiments


The CMS Data Acquisition

• O(104 ) distributed Objects to– control– configure– monitor

• On-line diagnostics and problem solving capability

• Highly interactive system (human reaction time - fraction of second)

• World Wide distributed monitor and control

2 107 electronics channels 40 MHz

100 Hz


CMS Prototype


CMS Prototype: IEs at work

Det 1Det 1Det 1

DAQ

TTS

FedBuilder RuBuilder

FilterFarm

Trigger

TOP

GTPe

DAQ

Detector1

8

- GridCC middleware used for CMS MTCC (Magnet Test and Cosmic Challenge)

- 11 Instrument Elements with a hierarchical topology

- Instruments are in these case Linux hosts where the cms on-line software is running

- More than 100 controlled hosts

- 25 days to the start of the data taking !

CMS Instrument Elements

DAQ IE Instrument Managers


1. Taking Control

Target domain

"zombies"

Pirated machinesDomain A

Pirated machinesDomain B

X

IDS Intrusion Detection System


A DDoS Attack Domain-wiseA DDoS Attack Domain-wise

Sensor Instrument Element

Target Domain

Sources of the attack





IDS Intrusion Detection System


Main GridCC Pilot Applications: Remote Operation of an Accelerator

Elettra Synchrotron


The other GridCC pilot applications• Meteorology (Ensemble Limited Area Forecasting)

• Device Farm for the Support of Cooperative Distributed Measurements in Telecommunications and Networking Laboratories

• Geo-hazards: Remote Operation of Geophysical Monitoring Network (see first slides)

• Medical Devices need a close loop between the data acquisition and the output result


Conclusion

• The GridCC project is integrating instrument into traditional computational/storage Grids.

• IEs need an high interaction and interactivity between itself and the users.

• The GridCC IE implementation is currently installed in heterogeneous applications


Question?

• Thx for your time

Acknowledgement: The GridCC project is supported under EU FP6 contract 511382.

More information: www.gridcc.orgOn-line Demo at: http://sadgw.lnl.infn.it:2002/IEFacade

http://sadgw.lnl.infn.it:2002/IEFacade

overview

Documents

remote control

gridcc projectit

distributed control

standard grid access

grid conceptssatellite

term instrument element

acquired information

orginstrument element