Web Tools, Mashups and Automated Lasers - University of Southampton

Introduction

The

introduction

of

software

based

laboratory

information

management

systems

(LIMS)

has

helped

manage

many

aspects

of

the

laboratory,

from

movement

of

samples

and

people

to

the

experiments

being

carried

out
.

These

management

systems

have

been

developed

further

to

interact

directly

with

devices

in

the

laboratory,

automating

certain

experiments

(such

as

auto
-
samplers)
.

The

use

of

online

systems

allows

for

remote

management

and

control

of

experiments

through

web

browsers

and,

where

possible,

through

hand

held

devices

such

as

mobile

phones
[
1
]
.

Allowing

remote

control

and

review

can

benefit

the

experimenter

in

a

number

of

ways

such

as

increasing

safety

where

the

experimental

environment

is

unsafe

to

remain

in,

improving

results

where

small

changes

in

the

environment

greatly

affect

the

obtained

results

and

can

save

time

through

early

detection

of

failures
.

Web Tools, Mashups and Automated Lasers


Stephen Wilson

&
Jeremy Frey*


School of Chemistry, University of Southampton,
Highfield
, Southampton, SO17
1BJ, UK;
sw1703@soton.ac.uk

Acknowledgments

Stephen

Wilson

thanks

the

EPSRC

for

a

studentship
.

Stephen

would

also

like

to

thank

Professor

Jeremy

Frey

and

the

research

group

for

their

support

in

the

project
.

References

1.
J.M
. Robinson, J.G. Frey, A.D. Reynolds, B.V.
Bedi
, and A.J. Stanford
-
Clark. From the laboratory to the mobile
phone:
Middleware
for laboratory data acquisition using the publication subscribe model. In e
-
Science 2005, 2005.

2.
Andy J. Stanford
-
Clark. Integrating monitoring and telemetry devices as part of enterprise information resources.
WebSphere

MQ Integrator, March
2002.

3.
Microsoft MSDN. Publish Subscribe. [web page] http://msdn.microsoft.com/en
-
us/library/ms978603.aspx. [Accessed
29 September 2009].

4.
MQ
-

Telemetry Transport.
MQTT.org
. [web page] http://mqtt.org/. [Accessed 29 September 2009].

5.
Simile. Simile. [web page] http://simile.mit.edu/. [Accessed 29 September 2009].

Fig. 1.
The interaction between producers and consumers using the message broker

Message

Brokering

In

this

project

we

have

developed

software

to

control

and

review

an

experiment

to

study

liquid

interfaces
.

The

data

flow,

of

both

captured

data

and

control

messages,

are

handled

through

a

centralised

message

broker
.

The

large

results

files

are

transferred

through

a

parallel

higher

throughput

system,

preventing

the

message

broker

from

becoming

overloaded
.

This

broker,

called

the

Microbroker,

is

part

of

the

IBM

WebSphere
[
2
]

software

suite
.

The

Microbroker

follows

the

publish/subscribe
[
3
]

architecture

for

controlling

the

flow

of

messages

using

the

message

queue

transport

telemetry

(MQTT)
[
4
]
.

A

data

producer

acquires

data

from

a

source

and

publishes

this

to

the

Microbroker
.

A

data

consumer

receives

message

for

which

they

are

subscribed

to
.

Using

the

message

broker

allows

the

software

to

be

independent

of

the

overall

system
;

such

that

should

a

new

data

source

be

introduced

no

changes

to

the

existing

software

are

required,

only

that

a

new

topic

will

be

available
.

To

include

a

new

data

consumer,

this

only

needs

to

subscribe

to

the

topics

it

is

interested

in,

rather

than

having

to

change

all

the

existing

data

producers
.

This

is

shown

in

Fig
.

1
.

Web

tools

The

development

of

Web

2
.
0

and

XML

has

led

to

a

number

of

tools

and

services

which

are

now

available

as

open

source

software
;

many

of

which

can

be

used

in

the

research

domain
.

Examples

of

these

can

be

well

known

websites

such

as

YouTube,

Flickr

or

more

specific

research

domain

tools

such

as

CiteULike
,

Connotea

or

Delicious
.

These

tools

are

scalable

to

a

number

of

domains

as

they

use

an

application

programming

interface

(API),

this

is

a

predefined

vocabulary

for

interacting

with

the

service

and

will

be

how

they

are

linked

to

the

message

broker
.

Although

these

tools

alone

are

useful,

many

can

be

combined

to

generate

a

more

descriptive

view

of

the

data
;

this

combination

of

web

services

and

tools

can

be

described

as

a

mashup
.

In

this

work

we

have

looked

at

tools

provided

by

the

Simile

group
[
5
]

such

as

graphing

and

facetted

searching

solutions

as

well

as

services

provided

by

Google
.

The

Liquid

Interface

Experiment

–

Control

In

this

project

an

experiment

to

study

the

liquid

interface

by

second

harmonic

generation

(SHG)

was

automated

using

the

Microbroker

to

control

the

flow

of

messages

both

to

the

apparatus

and

from

the

data

sources
.

The

SHG

experiment

consists

of

a

number

of

experimental

components,

each

of

which

can

be

controlled

through

RS
-
232
,

and

a

data

acquisition

card
.

The

main

parts

of

the

experimental

set

up

are

a

laser

source,

a

shutter,

two

polarisation

rotators

(input

and

output)

and

a

sample

diluter
;

the

experimental

set

up

is

shown

in

Fig
.

2
.

Simple

Perl

scripts

were

generated

to

subscribe

to

a

control

channel

of

the

message

broker,

when

a

message

is

received

this

was

converted

to

the

device

specific

command

and

sent

through

RS
-
232

to

the

device
.

As

the

message

broker

software

handles

all

the

communication

protocol

each

script

was

approximately

twenty

lines

of

code
.

At

this

time

security

is

handled

through

restricted

network

access

although

it

is

planned

to

implement

a

security

system

in

the

future
.

There

are

also

hardware

based

security

mechanisms

to

prevent

injury

and

damage
.

Fig.
2. The SHG laser apparatus showing the components of the system and how they communicate
with the message broker

The

Liquid

Interface

Experiment

–

Mashup

review

The

experimental

review

mashup

website

was

developed

in

PHP

and

uses

a

number

of

services

available

via

the

web

from

other

developers
.

To

allow

for

faster

searching

of

experiments

the

Simile

Timeline

software

is

used
.

The

Timeline

software

provides

an

overview

of

all

the

experiments

carried

out

via

a

timeline,

this

can

be

dragged

using

the

mouse

to

move

through

time
.

The

granularity

of

the

timeline

(such

as

hour,

day

or

year)

can

be

customised

and

combined

to

provide

faster

searching
.

Experiments

within

the

timeline

can

be

colour

coded

to

distinguish

between

certain

nodes
.

Once

an

experiment

has

been

selected

a

number

of

tools

are

used

to

generate

a

view

of

the

experiment

in

its

current

state
.

The

experimental

metadata

is

displayed

in

a

tabular

format

using

the

Google

Docs

API
;

here

each

cell

can

be

referenced

by

a

single

URI,

allowing

faster

abstraction

of

the

metadata
.

The

experimental

results

are

plotted

using

the

Google

Chart

API,

this

uses

a

single

URL

to

generate

a

plot
.

This

can

also

provide

mouse

over

functionality

to

display

information

about

each

point

and

embedded

links

within

the

plot

to

get

further

data

about

each

data

point
.

The

environmental

data

such

as

temperature,

humidity

and

light

levels

associated

with

the

experiment

can

also

be

reviewed
.

If

the

experiment

is

still

running

the

most

recent

values

are

shown

in

a

tabular

view
.

If

the

experiment

has

completed

the

Simile

Timeplot

software

is

used

to

generate

a

interactive

plot

of

this

environmental

data
.

The

Timeplot

software

reads

CSV

and

XML

data

files

and

generates

a

time

plot

of

the

data,

this

also

has

mouse

over

functionality
.

Fig
.

3
.

shows

a

number

of

screenshots

from

the

review

website

Fig.
3. Various parts of the SHG experimental mashup review website

Conclusions

The

use

of

the

Microbroker

as

a

standardised

communication

protocol

has

allowed

for

rapid

development

of

software

for

both

handling

acquired

data

and

sending

control

messages
.

As

each

component

is

only

required

to

interact

with

the

broker

the

system

can

be

set

up

as

a

distributed

deployment
.

This

allows

machine

demanding

processes,

such

as

data

acquisition,

to

run

on

a

remote

dedicated

server
;

reducing

the

load

on

the

other

machines
.

This

has

also

allowed

for

faster

integration

of

new

components,

the

addition

of

a

new

control

device

only

required

a

new

topic
.

The

use

of

parallel

transfer

systems

for

larger

data

files

has

prevented

the

broker

from

becoming

overloaded
.

It

has

been

shown

that

a

number

of

web

tools

and

services

can

be

used

in

the

research

domain

to

provide

a

detailed

view

of

experimental

data
.

The

ease

of

use

of

these

tools

is

greatly

increased

by

a

well

documented

API
.

When

combined,

these

tools

can

provide

more

detail

on

the

data

than

if

used

alone
.

The

Simile

tools

are

all

developed

with

inter
-
connectivity

in

mind

and

therefore

are

easy

to

integrate

with

each

other
.

Message
Broker

Data Producers

Data Consumers

Result Modelling

Result Review

Experiment Timeline

Data Point Review

Environmental conditions

Message Broker

Sample

Laser

source

Laser shutter

Polarisation

rotator

Polarisation

rotator

Detector

Dual syringe

diluter

Database