MICROSERVICES

Lightweight Service Descriptions for REST Architectural Style

Jos

e Ignacio Fern

andez-Villamor, Carlos

A. Iglesias and Mercedes Garijo

Departamento de Ingenier

ıa de Sistemas Telem

aticos, Universidad Polit

ecnica de Madrid, Spain

Keywords:

REST, Service description, Semantic web, Mashups, Web applications.

Abstract:

Current web has a vast number of applications available that offer users a wide domain of services. Most

services, however, cannot be machine processed, which limits service composition for application and mashup

development. Research on Semantic Web Services contributes to the improvement of interoperability and

composition of applications and services. Many approaches cover service description by following paradigms

such as Web Services and REST architectural style, allowing describing any kind of service for its use by an

automatic agent, but sometimes using these solutions can be a time-consuming task.

This paper introduces Microservices, a lightweight service classiﬁcation framework for REST architectural

style. Microservices do not attempt to describe every possible service, but to provide a way to describe a

set of services in a simple way. Microservice descriptions consist of a set of terms that represent service

features. After describing features semantically, microservices framework allows generating detailed service

descriptions, which allows reusing common feature descriptions across different services. A use case that

adapts heterogeneous search services to produce a standard interface using microservices is described.

1 INTRODUCTION

The web has an increasing number of applications

and services that cover different domains and ﬁelds.

Users can enjoy a wide range of applications, from

e-commerce to blogging, media or social network-

ing. The possibilities of the current web are only lim-

ited by the interoperability between applications. In-

ternet versatility would increase if applications could

be arbitrarily composed and automatically executed

to fulﬁl a user’s goal. This approach of building a

programmable web has led to research in several ar-

eas, such as semantic web services (Zhou et al., 2006)

or mashups (Yu et al., 2008). Many approaches have

been successful from a research point of view without

having reached wide adoption. The effort of describ-

ing services in web applications can be too big and

too often web developers do not perform that task.

This paper introduces microservices, a framework

that attempts to simplify service description, and its

application on a use case. With microservices frame-

work, service descriptions can be generated out of

simple service feature lists, enabling services in web

applications to be consumed by automatic agents.

This paper is organized as follows. First, a review

of previous related approaches is done and the back-

ground of the problem is summarized. Second, our

approach to service description is described. Third, a

use case that employs microservices to adapt hetero-

geneous search services is outlined. Finally, the main

conclusions of this research work are summarized.

2 BACKGROUND

Current web has applications with plenty of services

available, which are of many different kinds. Ap-

proaches to describe services semantically allow au-

tomatic agents to execute and compose services au-

tomatically. Some approaches follow the Web Ser-

vices Architecture, such as OWL-S (World Wide Web

Consortium, 2004), WSMO (Roman et al., 2005),

METEOR-S (Patil et al., 2004) or WSDL-S (World

Wide Web Consortium, 2005). To favour the inte-

gration with Internet’s architecture, RESTful services

started to be employed in web applications, which

caused RESTful alternatives to semantic web services

to appear. SA-REST (Sheth et al., 2007) or hRESTs

(Wright State University, 2008) are approaches that

provide languages to describe RESTful APIs speciﬁ-

cations. Similarly, Web Application Description Lan-

576

Ignacio Fernández-Villamor J., Á. Iglesias C. and Garijo M. (2010).

MICROSERVICES - Lightweight Service Descriptions for REST Architectural Style.

In Proceedings of the 2nd International Conference on Agents and Artiﬁcial Intelligence - Artiﬁcial Intelligence, pages 576-579

DOI: 10.5220/0002720105760579

 SciTePress

Figure 1: Microservice description framework.

guage (Hadley, 2006) proposes describing RESTful

APIS by deﬁning a WADL, XML-based, ﬁle.

All these approaches provide means to describe

every kind of service and process the respective de-

scriptions to automate tasks such as execution or com-

position. They are abstract and ﬂexible enough to al-

low describing every possible service. Whenever a

new service is deployed, a description has to be built

in order to make the service available for automatic

agents to process it. Usually, this implies describing

service’s inputs and outputs semantically, as well as

deﬁning the precise operation the service performs,

which can be a time-consuming task that is sometimes

not carried out.

3 MICROSERVICES

Microservices framework attempts to simplify the

process of deﬁning service descriptions to push auto-

matic service consumption in the semantic web. In

this framework, the description task attempts to be

improved by enabling reusability accross service de-

scriptions.

A microservice description is a list of terms. Each

term represents a feature that the service has. Exam-

ples of features can be “the service performs a re-

trieval operation”, “the service requires user authen-

tication”, “the service performs a storage operation”,

“the service has an image as input”, or “the service

outputs a set of resources”. By describing features

semantically, feature descriptions can be combined to

produce a semantic service description. This allows

reusing feature descriptions in services that are differ-

ent but have one or more features in common. This

feature-based approach is shown in ﬁgure 1, and has

four elements of interest:

• A microservice description, which consists of a

list of terms. Each term represents a feature that

the service has.

• An extended service description, which is deﬁned

as a textual description and a set of preconditions

and postconditions that deﬁne the operation that

is performed by the service.

• A transformation function that transforms a mi-

croservice description into an extended service

description.

• A set of term deﬁnitions that deﬁne the semantics

of terms and thus describe service features.

An example of microservice description is

keyword-filtered multiple picture get,

which describes a search service of pictures that are

ﬁltered by keywords. Four terms are used in order

to deﬁne the microservice: keyword-filtered,

multiple, picture, and get. Because of following

REST architectural style, at least a term that repre-

sents the underlying HTTP method used has to be

included in the description.

An extended service description is built by com-

bining the appropriate term deﬁnitions. Deﬁnitions

can be tied to more than one term. For example, there

can be a deﬁnition for get, a deﬁnition for picture,

but also a deﬁnition for altogether picture and get.

The need for deﬁnitions that involve sets of terms

can be easily noticed. Term picture should set a

postcondition (i.e. “outputting a picture”) when used

with term get, but should set a precondition (i.e. “a

picture has to be provided as parameter”) when used

with term post. This can be achieved by setting def-

initions for picture and get and for picture and

post.

In this case, the extended microservice description

for the previous example is built by combining the

deﬁnition for keyword-filtered, the deﬁnition for

multiple and get, the deﬁnition for picture and

get, and the deﬁnition for get by grouping their re-

spective preconditions, postconditions and textual de-

scriptions. This service’s extended description can be

seen in ﬁgure 2.

As a feature-based description framework, the mi-

croservice approach has several advantages over man-

ually deﬁning a service description: (i) it allows the

reuse of feature descriptions among different services,

and (ii) it reduces the description task to selecting a

set of features that describes the considered service,

given a vocabulary of terms.

4 USE CASE: DESCRIBING

SEARCH SERVICES

An environment for the creation of microservices

through the deﬁnition of microservice proxies to other

MICROSERVICES - Lightweight Service Descriptions for REST Architectural Style

577

Figure 2: Sample use case scenario.

external services has been developed

. This envi-

ronment allows wrapping an already existing service

into a proxy that exposes an interface that can be de-

scribed with a microservice description. Microservice

proxies are deﬁned by specifying a preadapter, which

adapts service inputs, and a postadapter, which adapts

the output format. This environment is used in this

use case to wrap Google and Flickr search services.

In order to describe the proxies to these services,

terms keyword-filtered, picture, multiple, and

get, and their associated term deﬁnitions as shown in

table 1, are deﬁned.

The result of wrapping these search services as

microservice proxies, as shown in ﬁgure 2, is:

• The described microservices share most of their

terms in their descriptions, which allows reutiliza-

tion of the semantic descriptions.

• Both services share the same interface. Sam-

ple searches can be tried at Google’s proxy

and

Flickr’s proxy

. All of them accept a set of key-

words in input keywords and have the same out-

put format. As Flickr’s proxy searches pictures, it

additionally includes an image ﬁeld in its results.

• Services are advertised in an HTML form at a

URI. This HTML form allows the execution of the

service by a human user.

http://lab.gsi.dit.upm.es/microservices

http://lab.gsi.dit.upm.es/microservices/

http://www.google.com/search?keywords=test

http://lab.gsi.dit.upm.es/microservices/

http://www.ﬂickr.com/search?keywords=test

• Automated validation. Preconditions and post-

conditions are checked when executing a service,

which allows checking the service’s correct exe-

cution.

• Automated discovery. Microservice descriptions

are advertised as Linked Data in the HTML form,

which allows processing by an automatic agent.

5 RELATED WORK

Some ongoing research works focus on HTML form

description. RDForms (Baker, 2005) attempt to “add

to the Semantic Web capabilities similar to HTML

forms”. Schemas for indexable, container and set-

table operations are deﬁned, which represent HTTP

GET, POST, and PUT methods, respectively.

OpenSearch (A9.com, inc., 2005) is an approach

to modelling search services. It is aimed at a very

speciﬁc kind of service, which allows building ﬁne-

grained speciﬁcations.

However, these and other already mentioned al-

ternatives such as WSMO, SA-REST or WADL are

heavy-weight approaches that provide means to de-

scribe every possible service from scratch. On the

contrary, microservices framework attempts to allow

reusing other service’s descriptions.

ICAART 2010 - 2nd International Conference on Agents and Artificial Intelligence

578

Table 1: Term deﬁnitions for search services.

Terms picture, get

Description Output representation is a picture

Postconditions output.resources.checkAll(’i.image != null’)

Terms keyword-filtered

Description Keywords are used to ﬁlter the output

Preconditions params.keywords

params.keywords.length > 0

Postconditions output.resources.checkAll(’i.keywords != null’)

output.resources.checkAll(’i.keywords.contains(params.keywords.split(" "))’)

Terms get

Description Executing this service will retrieve a representation given by the Request-URI

Postconditions response.status == 200

Terms multiple, get

Description Multiple resources may be returned in the form of a list

Postconditions output.resources.length() >= 1

6 CONCLUSIONS

In this paper, current service automation approaches

have been reviewed. In general, these approaches are

focused on providing ﬂexible ways to describe every

possible service. However, this ﬂexibility occasion-

ally makes service description a time-consuming task.

In this paper, we have introduced microservices

as an alternative to simplify service description. Mi-

croservice descriptions are simple and allow automat-

ing various tasks in order to push service automation

in the semantic web. A use case that adapts a set

of heterogeneous search services has been described,

offering a uniform interface, discoverability, and au-

tomating documentation and validation.

Future work involves analyzing other case studies

and considering automatic wrapper induction to sim-

plify the task of building microservice proxies.

ACKNOWLEDGEMENTS

This research project is funded by the European Com-

mission under the R&D project ROMULUS (FP7-

ICT-2007-1) and by the Spanish Government under

the R&D project Java sobre Ruedas (FIT-350401-

2007-8).

REFERENCES

A9.com, inc. (2005). OpenSearch speciﬁcation.

http://www.opensearch.org/Speciﬁcations/Open-

Search/1.1.

Baker, M. (2005). RDF Forms. http://www.markbaker.ca/

2003/05/RDF-Forms/.

Hadley, M. J. (2006). Web application description language.

https://wadl.dev.java.net/wadl20061109.pdf.

Patil, A., Oundhakar, S., Sheth, A., and Verma, K. (2004).

METEOR-S Web service Annotation Framework. In

Proceeding of the World Wide Web Conference.

Roman, D., Keller, U., Lausen, H., de Bruijn, J., Lara, R.,

Stollberg, M., Polleres, A., Feier, C., Bussler, C., and

Fensel, D. (2005). Web Service Modeling Ontology,

Applied Ontology. IOS Press.

Sheth, A. P., Gomadam, K., and Lathem, J. (2007). SA-

REST: Semantically Interoperable and Easier-to-Use

Services and Mashups. In IEEE Computer Society.

World Wide Web Consortium (2004). OWL-

S: Semantic Markup for Web Services.

http://www.w3.org/Submission/OWL-S/.

World Wide Web Consortium (2005). Web Ser-

vice Semantics - WSDL-S. http://www.w3.org/

Submission/WSDL-S/.

Wright State University (2008). HTML Microfor-

mat for Describing RESTful Web Services and

APIs. http://knoesis.wright.edu/research/srl/projects/

hRESTs/#hRESTs.

Yu, J., Benatallah, B., Casati, F., and Daniel, F. (2008).

Understanding mashup development. IEEE Internet

Computing, 12(5):44–52.

Zhou, J., Koivisto, J.-P., and Niemela, E. (2006). A survey

on semantic web services and a case study. In Com-

puter Supported Cooperative Work in Design, 2006.

CSCWD ’06. 10th International Conference on, pages

1–7.

MICROSERVICES - Lightweight Service Descriptions for REST Architectural Style

579