=Paper=
{{Paper
|id=Vol-1871/paper1
|storemode=property
|title=A Model for Information and Action Flows Connecting Science Gateways to Distributed Computing Infrastructures
|pdfUrl=https://ceur-ws.org/Vol-1871/paper1.pdf
|volume=Vol-1871
|authors=Gabriele Pierantoni,Dermot Frost,Sandra Gesing,Silvia D. Olabarriaga,Mahdi Jaghoori,Gabor Terstyanszky,Junaid Arshad
|dblpUrl=https://dblp.org/rec/conf/iwsg/PierantoniFGOJT16
}}
==A Model for Information and Action Flows Connecting Science Gateways to Distributed Computing Infrastructures==
https://ceur-ws.org/Vol-1871/paper1.pdf
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
A Model for Information and Action Flows
Connecting Science Gateways to Distributed
Computing Infrastructures
G. Pierantoni∗ , D. Frost∗ , S. Gesing† , S. Olabarriaga‡ , M. Jaghoori‡ , G. Terstyanszky§ and J. Arshaid§
∗ Trinity Center for High Performance Computing, Trinity College Dublin, Dublin, Ireland
Email: pierantg@tcd.ie
† Center for Research Computing, Department of Computer Science and Engineering, University of Notre Dame, IN, USA
‡ Academic Medical Center of the University of Amsterdam, The Netherlands
§ Center for Parallel Computing, University of Westminster, London, United Kingdom
Abstract—To support scientists of different disciplines, differ- many successful efforts [3], [5]–[8]. These resulted in the
ent fields of Computer Science have developed tools and infras- construction of abstraction layers capable of interfacing with
tructures with the aim of giving them access to vast computational heterogeneous, distributed systems in a unified fashion.
resources in the easiest possible way. Such extremely complex
structures have evolved naturally in the last decades both in depth The need to formalize and share the scientific process
and breath and, in addition to scientists, a plethora of heteroge- have also been satisfied by different scientific communities by
neous actors (system administrators, developers, etc.) cooperate adopting the workflows concept originally developed for the
and interact with them. This complex and unstructured flow of industry. Several such workflows [9]–[12] have been developed
actions and information poses difficulties in the development and and have been adopted by different scientific communities,
usage of Science Gateways because information can be missing
or hard to isolate at the right layer. In this paper, we aim to start giving raise to the same interoperability problem as found in
a discussion on how to best manage these information flows to distributed computing. The workflows interoperability prob-
help the design and implementation of more flexible and user- lem [13]–[15] has been addressed by building abstraction
friendly Science Gateways and workflow management systems in layers and intermediate languages. Nevertheless, while such
the future. efforts aim at a relative degree of freedom and interoperability
Index Terms—Workflows, eScience Portals, eInfrastructures,
Science Gateways, Information Flows, Interoperability, Dis- across different Workflow Management Systems, they also
tributed Computing Infrastructures, Workflow Management Sys- increase the complexity of the information flows.
tems All these layers, of infrastructure and workflows, are con-
nected to each other by flows of requests and replies that are
I. I NTRODUCTION unstructured and heterogeneous by nature. Requests propagate
In modern days, science relies on computation to such an downward from the upper to the lower layers, while replies
extent that the term in silico has been added to the terms in propagate upward from the lower to the upper layers and
vivo and in vitro. To support scientists of different disciplines eventually reach the users who originated them. Replies carry
in accessing computational resources that are ever growing in information on the status and on the outcome of the request
size and complexity, different fields of Computational Science (often merged together), offering to the upper layer a partial
have developed tools and infrastructures that fall under the view on the overall information of the lower layer.
broad definition of Science Gateways (SGs) [1]–[3]. Science These multi-layered infrastructures are used by a plethora
Gateways lie on the top of extremely complex systems and of actors with different skills [4], inclinations and priorities,
services that have evolved naturally in the last decades both which increase the complexity to a higher level. Administra-
in depth and breath. They span multiple layers specialized tors, developers, and scientists, all of them interact with one
in tackling specific facets of the challenge and different or more layers, and each of them is interested in a subset of
communities have developed independent implementations for the information flow in each layer with which she/he is likely
each layer. Furthermore, in addition to scientists, a plethora to be best acquainted.
of heterogeneous actors (system administrators, developers, The complexity of these information flows poses relevant
etc.) [4] cooperate with the scientists and interact with the difficulties in the development and usage of Science Gateways,
infrastructure. The challenges are enourmous to make all as information can be missing or hard to isolate at the right
systems and persons communicate and interoperate. layer. This is true both for scientific users and administrators
The challenges posed by the need to harness distributed (e.g. error messages can be absent or difficult to understand),
computing infrastructures (DCI) that vary greatly in their but also for developers as it is difficult to build systems that
implementation, such as Clouds, Grids, Desktop Grids and autonomously react to undesired events, and to dispatch the
High Performance Computing, have been at the center of right information type to the proper users.
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
In this paper we start a discussion on how to best manage Information Domain; we also define Information flows as the
these information flows to help the design and implementation exchange of information between Elements in different Layers.
of more flexible and user-friendly Science Gateways and The information flows that describe part of the Information
Workflow Management Systems in the future. The first step Domains to the upper layers have different characteristics
of this discussion is to propose a model to describe these and related challenges. We focus here on three such chal-
information flows and the architecture that hosts them. The lenges: Heterogeneous information represents the challenge
domain we are attempting to model is extremely vast and of high utility information mixed with information of less
diverse, so we start by analyzing a sub-domain encompassing relevance. Incomplete information represents the challenge that
solutions with which the authors are well acquainted. Also, arises when Users cannot directly access all the required in-
this paper proposes a qualitative approach without any formal formation. Finally, the problem ofInformation interoperability
description that will be attempted after the initial model has arises whenever different implementation of the same Layer
been validated. The overall goal is to increase the usability impose the use of different languages and interaction patterns
for the diverse user groups of Science Gateway systems. Our to perform the same action.
model considers standards, setting the context and suggesting Also, from each user’s perspective, information flows may
methods for measuring user experience. Such standards in- be more or less useful and more or less easy to manipulate. To
clude CISU-R (Common Industry Specification for Usability describe this, we introduce the concepts of Utility, Cost and
Requirements) [16] developed by the Visualization and Us- Value. Utility defines the usefulness to the user, Cost describes
ability Group within NIST (National Institute of Standards the difficulty to obtain the information and Value represents
and Technology) [17]. We start with on initial level with an the difference between the two.
expert evaluation [18], which is based on our own knowledge, To increase the overall Value of the information, we have
experience and use cases. Since we are developers, providers observed that the scientific communities have devised different
and also users of science gateways and workflow systems, such systems. Heterogeneity Reduction Functions do not modify
an expert evaluation covers already a broad view. the Utility of information but reduce the Cost associated
The paper is structured as follows: Section II introduces to their fruition. Information Extension Functions increment
preliminary concepts and terminology, and Section III intro- the Utility of information while maintaining its Cost fixed.
duces the model to describe the information flows. Section IV Interoperability Functions offer a unified interface to multiple
describes some currently used tools and technologies and implementations of the same Layer.
Section V discusses the road ahead.
mds May 10, 2016 III. A M ODEL FOR I NFORMATION F LOWS
The domain we attempt to model spans multiple layers and
II. C ONCEPTS AND T ERMINOLOGY many different implementations for each layer, therefore it is
therefore arduous to draw a conclusive and exhaustive schema.
The proposed model is based on the followig assumptions
Nevertheless, we observed some recurring architectural pat-
and concepts. Firstly, we assume that Science Gateways are
terns that suggest to adopt an abstraction encompassing four
composed of several Layers. We define a Layer as an entity
main layers:
that represents an element of a Science Gateway. A Layer can
• Scientific Domain layer for interaction with the scientific
have different implementations and it exposes a well-defined
set of functionalities to its users. Example of Layers are user using domain concepts.
• Generic Portal layer for interaction with generic users
Presentation and Service Layers, Distributed Computing In-
frastructure Layers (e.g. Grids and Clouds), Workflow Layers and to offer tools and APIs to build the applications of
(e.g. TAVERNA or WS-PGRADE). Each Layer is composed the above layer.
• Workflow Management layer, where the processing or-
of various Elements and is described by its Status
Elements of each Layer fall into two main categories: chestration is described and executed.
• Distributed Infrastructures used for computing, storage
structural and transient. Structural Elements are static entities
that deliver functions and services inside the Layer (e.g. the and data, which are normally represented by one or more
Job Execution Service or Information Service in a DCI). DCIs.
Transient Elements are dynamic entities created by the user To model how the different layers and actors interact
inside the Layer (e.g. jobs description, files, workflows) to through Information Flows, we try to simplify such a complex
run a specific application. Each Layer can be accessed by system and then adapt step by step the model to the complexity
Access Components, which are entities that enable access. of real systems.
Access Components can be user-oriented such as Graphical The layers are examined under the consideration of CISU-
User Interfaces and Command Line Tools or programmatic R, which defines three levels of compliance for usability. Level
interfaces such as an API. 1: Context of use must consider individually:
Layers communicate through the means of Requests • The stakeholders.
and Replies and are defined by the status of their Structural • The intended user groups.
and Transient Elements. We define their combination as its • The main goals for each user group.
2
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
• The intended computing or technical environment. • Structural Elements model the internal components of the
• The intended physical and social environments. layer.
• Scenarios of use specifying tasks in context. • Transient Elements model the objects defined by the
• Any prerequisite documentation/training materials. user that are currently handled by the layer (e.g jobs or
Level 2: Measures must include: Workflows being managed).
• Performance measures, i.e. achieving user goals. External to the layer, there are either human users or other pro-
• Satisfaction measures via known questionnaires. grammatic entities that connect to it. We model the interactions
Level 3: The test method specifies how it is planned to evaluate between these entities as represented in Figure 1 by employing
that the requirements are met. the concepts of Layers, Requests, Replies, Structural Elements,
The model focuses at this stage on setting the context and Transient Elements.
regarding level 1: from stakeholders (science gateway and Layer I+1 issues Requests (possibly involving Transient
workflow management providers), intended user groups (e.g., Elements) to the Layer I and obtains Replies in return. It is
domain scientists, administrators) and main goals for each important to highlight that an action may modify the status of
group. The computing and technical environment as well as both Structural and Transient Elements of the layer, but the
scenarios are analyzed via case studies. Keeping the model as ending status of these entities does not strictly follow the ones
generic as possible, we aim to incorporate and apply it for that preceded the action (as other events may have occurred
diverse physical and social environments. while the action was executed). Requests are detailed by
parameters that may include Transient Elements or references
A. The Model of one Layer to them. As an example, the submission of a job to a DCI
Here we model a single layer of the full stack and its can be modeled as a Request of a submission action of a
interactions with the users. It is important to stress that this Transient Element describing a job that will take as parameters
does not attempt to model an entire stack as a Single Layer the job description itself, additional parameters and details of
but rather to model a generic layer of the full stack. the identity of the entity submitting the job. Replies include
Figure 1 presents a simple model of a generic layer of the different, heterogeneous elements such as an exit/error code,
stack, which could be used to model a portal, a workflow job results and logging information.
submission system or a DCI middleware. We also model the situation when users may not be able
allowed to issue all Requests to a layer and that they may
not be able to directly access the entire set of the Information
Domain of the Layer. This can be the case of Authorization
Layer I+1 policies. We define the subset of the Information Domain
User
accessible by each user as being Directly Accessible.
Request Domain Reply Domain
Finally, we model the different profiles of actors con-
(R) (P) necting to the layer through three main profiles: Result-
Access to Layer I
Oriented, Layer-Oriented, and Development-Oriented actors.
Structurual Transient Henceforth we will refer to all actors accessing Layer I
Elements Layer I Elements
S S E E E as Users encompassing in this generic term both human users
Information Domain
and programmatic components. In any case, even software
(I) components will have to be executed with a certain identity
either by delegation, robot certificates or other means. Result-
Fig. 1. Model of a Single Layer oriented users model actors whose main interest is in the
results provided by the layer. They want to be shielded as much
The proposed model comprises the following entities: as possible from the technical details of the layer. Ideally, a
• Layer I describes a generic Layer in the structure such Result-Oriented User would like to treat the entire layer as a
as a portal, a Workflow Management System, or DCI). black box that would either return the results correctly or, in
Layer I will be described by its Information Domain that case of failure, deliver within expected time the result along
includes its status. The description of the status has to with a contact point for addressing the issue. Since jobs or
take into account the dual, interconnected nature of the tasks can be active over long periods of time, it is important
Layer: that of its own structure and that of the actions it to provide and visualize information for monitoring active
is performing. An example of this is the possibility of a jobs. Layer-oriented users model actors that have an opposite
job to fail because of the inconsistent status of the DCI view. They are interested in the internals and status of the
or because of a failure of the job itself. layer, which should be seen as a transparent box allowing
• Access to Layer I is an Access Component that models complete access and manipulation of the inner workings. They
APIs for programmatic access as well as command line are mainly concerned with the maintenance of the Structural
and graphical user interfaces for direct human interaction. Elements of the layer. Such users include Workflow Man-
Access Components can restrict access to the Layer agement and DCI Providers, who need detailed information
depending on Authorization policies. optimization and error resolution. Development-Oriented users
3
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
model actors whose main focus is the development of either languages and interaction patterns to perform the same action
the Structural Elements that compose layers or Transient (e.g. the execution of a workflows), which we coin Information
Elements such as Workflows on behalf of Result-Oriented Interoperability.
users. Science Gateway developers have devised different solu-
We also have to model the fact that layers have multiple im- tions to these problems which we attempt to model as ei-
plementations as presented in 2. In this case Layer I+1 triggers ther a Structural Element-Value Increasing Structural Element
actions and receives results from two separate implementations (VISE), or as a smart Transient Element - Value Increasing
of the same model. As there is no explicit interoperability Transient Element (VITE). VISE’s are usually result of the
provision between the two implementations, Layer I+1 will effort of Layer-Oriented Users that modify Structural Elements
have to support two separate access modalities (syntax that of one Layer to increase its usability by one or more Users.
defines Requests, Transient Elements, Reply formats, etc...) VISE’s are usually the result of efforts by Development-
by explicitly dealing with two separate Requests, Replies and Oriented or Result-Oriented Users that modify job descriptions
Information Domains. or workflows to increase the usability of one Layer, for
example jobs or Workflows that internally manage information
flows and/or automatically perform value-adding actions.
Layer I+1
User
Request Domain (Rx) Reply Domain (Px) Request Domain (Ry) Reply Domain (Py) Value Increasing Components Value Increasing Elements
(Implementation X) (Implementation X) (Implementation Y) (Implementation Y) (VIC) (VIE)
Access to Layer I Access to Layer I
(Implementation X) (Implementation Y)
Layer I Layer I
Information Domain Ix Information Domain Iy Layer I+1
(Implementation X) (Implementation Y) User
Layer I+1
User
Request Domain Reply Domain
Fig. 2. Model of a Single Layer with Multiple Implementations (R*) (P*)
Request Domain Reply Domain
VISE (R and R*) (P and P*)
Information Domain (I*)
B. Information Value Layer I
Request Domain Reply Domain
The concepts of Value, Utility and Cost are fundamental (R) (P)
I E I E
in this model, which can be expressed differently for each of Information Domain (I and I*)
the user Profiles. Utility describes how useful the information Layer I
Structural Transient
Information Domain (I)
contained within a Reply is to any particular user profile. The VIEs VIEs
Cost describes how difficult it is to obtain that information,
thus covering both the action of extracting information from
the Reply and issuing the related Request (e.g. extracting the
Fig. 3. Value Increasing Components and Elements
relevant information about the failure of a Workflows executed
on multiple DCIs may be very hard to perform). Finally Value
describes the difference between the Utility and the Cost. We envisage four main types of operations to increase the
Utility, Cost and Value need to be quantified for usage in value of Information Flows, the so called .
the proposed model, which is still topic of ongoing debate. A
1) Heterogeneity Reduction Function: They are filtering
possibility would be to use a real numbers in the range from
functions that isolate sub-set of a Reply in order to make
0 to 1, where 0 represents low scores. For example, a Request
it more accessible to different user profiles. An example
returning useless information (Utility=0) that is very difficult
would be a function that filters job results, error codes and
to understand for a particular user profile (Cost=1) would have
logging information. Result-Oriented Users will see only the
a Value of -1. Another Request returning useful information
job result, if any, while Layer-Oriented Users will see only
(Utility=1) that is very easy to understand (Cost=0) would
error codes and logging information, if directly available.
have a have a Value of +1.
These functions do not increase Utility but reduce the Cost
C. Value Increasing Functions thus increasing Value.
There are three main characteristics that reduce the Value 2) Information Extension Function: The second type ex-
to different user profiles. Firstly the information that has the tends the information domain directly available to the user. An
highest Utility to different User profiles is often mixed with example would be a an automatic operation that automatically
information that has less Utility. We refer to this problem retrieves information on the status of the Layer (e.g. retrieval
as that of Heterogeneous Information. Secondly, the required and parsing of log files) on the failure of a job. These functions
information may not be directly reachable by a user, a problem do not modify the status of the Transient Elements in the layer,
that we refer to as Incomplete Information. Finally, different but the Information Domain made available to the user has a
implementations of the same layers impose the use of different greater Utility and the same Cost thus resulting in increased
4
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
Value1 .
VISE as a layer VISE as part of the access to layer VISE as internal component
3) Compound Actions: The third type of functions, coined
Compound Actions, can perform a variety of actions that can
modify the status of the Transient Elements in the Layer. An
example would be the execution of pilot jobs prior to the
submission of the real job to foresee problems in the layer
and/or the automatic execution of diagnostic jobs and routines
on the failure of jobs.
4) Interoperability Actions and Functions: The final type
of functions, which we coin Interoperability Functions offer a
unified interface to multiple implementations of the same layer.
Examples are the submission to multiple DCIs infrastructures Fig. 4. Different ways to connect Layers with Value Increasing Structural
Elements
by a single set of commands or the possibility to execute a
workflow written for one Workflow Management System on a
different system. These functions offer the combined Utility of
Multi-Layer VISE Multi-Layer and Multi-Implementation VISE
several implementation while requesting the same Cost of one
implementation, thus radically increasing the overall Value
D. Interactions between VISEs, VITEs and Layers
We argue that there are five different ways in which Layers
and VISEs can interact. We also define a set of Improved Re-
quests, Improved Replies and Improved Information Domains
provided by the VISE.
• The first solution is to use a VISE as an additional,
separate Layer as presented in the left part of Fig-
ure 4. In this case the upper Layer can access both
Improved and Original sets of Requests and Replies.
This arrangement supports VISEs that offer abstractions
of different implementations of the same Layer allowing Fig. 5. Multi-Layer and Multi-Implementation VISEs
for interoperability. Examples are a VISE that optimizes
the Information Flow of different DCIs or a VISE that
optimizes the information flows from different Workflow SHIWA [20] and ER-FLOW [21]). These projects brought
Management Systems together domain experts and technology providers to work on
• The central part of Figure 4 shows a VISE embedded
a platform centered on the gUSE/WS-PGrade technology [22]
within the access component of Layer I to lower the complexity in the use of multi-layered infras-
• The right part of Figure 4 shows an opposite solution
tructures and many of these solutions can be seen as Value
where a VISE is embedded in Layer I+1 that uses it in Increasing Structural Elements (VISE’s) and Value Increasing
a totally transparent way Transient Elements (VITEs). From a broad perspective, the
• Figure 5 shows a Multi-Layer VISE that connects to
entire gUSE/WS-PGrade/SHIWA suite of components is a
different layers on the left and a Multi-Layer Multi- framework that can be used directly as a general Science Gate-
Implementation VISE on the right. A multi-layer VISE way to serve multiple scientific communities (Layer-Oriented
has the advantage of offering improved Requests and and Result-Oriented users) or can be used by Developer-
Replies that combine the Information Domains of mul- Oriented Users to create Customized Gateways that act as
tiple layers. A Multi-Layer, Multi-Implementation VISE Value Increasing Components built in the topmost layer for
extends the functionalities of a Multi-Layer VISE across Result-Oriented Users of specific communities.
different implementations of the same layer. Interoperability at computation infrastructure level is sup-
ported by the gUSE technology, which offers an abstraction
IV. E XAMPLES layer to multiple heterogeneous providers (Grid, HPC re-
sources, Desktop grids) called DCI-Bridge. It can be combined
The examples presented here represent the experience from
with another VISE offering an abstraction layer to different
a large of Science Gateway-oriented solutions of knowledge
Cloud Providers called CloudBroker [23]. A VISE offering
to the authors.
data transfer compound action across heterogeneous storage
Interoperability has been tackled at both the Workflow and
systems called DataAvenue [24] can also be connected to
DCI layers within a set of related projects (SCI-BUS [19],
gUSE to support domain experts who need to manage large
1 The status of the Layer and its Transient Elements can be modified in data sets across multiple sites.
subtle ways that are not covered at the moment by our model Interoperability between different implementation of Work-
5
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
flow systems has been achieved by the SHIWA Interoperability ACKNOWLEDGMENT
platform [14], a VISE that acts as an abstraction layer to
different Workflow Management Systems. Issues related at This research would not have been possible without the help
the reduction of expressiveness and richness of the set of and support of all the partners of the ER-FLOW, SCI-BUS
actions posed by the abstraction layer have been investigated and HELIO projects. The authors are particularly grateful to
and experimented with by four different communities in the Prof. Kacsuk, Prof Herres-Pawlis, Dr. Sciacca, Dr. Becciani,
ER-FLOW project [21]. Another example is presented by [7] Dr. Castelli and Dr. Taffoni.
to tackle the problem of Interoperability of Workflows across
heterogeneous infrastructures: Workflows are decorated with R EFERENCES
extra nodes to orchestrate the creation and destruction of
[1] N. Wilkins-Diehr, “Special issue: Science gateways - Common
Virtual Infrastructures that are enacted by VISEs operating community interfaces to grid resources,” Concurrency Computation
in Clouds. Practice and Experience, vol. 19, no. 6, pp. 743–749, apr 2007.
Interestingly enough, the solutions fall under two main [Online]. Available: http://doi.wiley.com/10.1002/cpe.1098
[2] P. Kacsuk, Science gateways for distributed computing infrastructures:
categories. The technology-oriented partners in SCI-BUS com- Development framework and exploitation by scientific user communities,
posed by Layer-Oriented and Development-Oriented experts P. Kacsuk, Ed. Cham: Springer International Publishing, 2014. [Online].
created VISE’s that had structural relevance in the overall Available: http://link.springer.com/10.1007/978-3-319-11268-8
[3] S. Gesing, T. Kiss, and G. Pierantoni, “Editorial:
architecture of the system. The domain expert partners com- Science Gateways Applying Clouds, Grids and
posed mainly by Result-Oriented and Development-Oriented HPC - 5819z007.pdf,” 2014. [Online]. Available:
experts developed multi-layered design patterns for their own http://www.computer.org/csdl/proceedings/iwsg/2014/5819/00/5819z007.pdf
[4] S. Olabarriaga, G. Pierantoni, G. Taffoni, E. Sciacca, M. Jaghoori,
Workflows that acted as Value Increasing Transient Elements V. Korkhov, G. Castelli, C. Vuerli, U. Becciani, E. Carley,
(VITEs). Three out of four communities in ER-FLOW devel- and B. Bentley, “Scientific Workflow Management – For
oped Value-Increasing Transient Elements by creating multi- Whom?” in 2014 IEEE 10th International Conference on
e-Science, vol. 1. IEEE, oct 2014, pp. 298–305. [Online]. Available:
layered patterns of Workflows [25]–[27] that exposed a higher http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=6972277
Value to the user than their individual parts and performed [5] K. Karoczkai, A. Kertesz, and P. Kacsuk, “Brokering Solution
compound actions when executed. One community took a for Science Gateways Using Multiple Distributed Computing
Infrastructures,” in 2015 7th International Workshop on
more general approach by developing a VISE, called Process- Science Gateways. IEEE, jun 2015, pp. 28–33. [Online]. Available:
ing Manager [8], that offered compound actions and increasing http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7217925
the Value of the information to the user by implementing an [6] Z. Farkas, P. Kacsuk, and A. Hajnal, “Connecting
abstraction layer to different levels of middleware. Workflow-Oriented Science Gateways to Multi-cloud
Systems,” in 2015 7th International Workshop on Science
One of communities used a powerful feature offered by Gateways. IEEE, jun 2015, pp. 40–46. [Online]. Available:
gUSE that allowed the easy development of interfaces [28] http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7217927
specific for each workflow. This solution as VISE in the [7] P. Kacsuk, G. Kecskemeti, A. Kertesz, Z. Nemeth, A. Visegradi, and
M. Gergely, “Infrastructure Aware Scientific Workflows and Their
upper layer, by selecting the set of information relevant to Support by a Science Gateway,” in 2015 7th International Workshop on
Result-Oriented actors, thus combining the rapidity offered Science Gateways. IEEE, jun 2015, pp. 22–27. [Online]. Available:
by template-driven development with the effectiveness of in- http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7217924
[8] M. M. Jaghoori, S. Shahand, and S. D. Olabarriaga, “Processing
formation heterogeneity-reduction actions. This solution acted Manager for Science Gateways,” in 2015 7th International Workshop
on three different aspects: First, it reduced the Cost of issu- on Science Gateways. IEEE, jun 2015, pp. 1–7. [Online]. Available:
ing requests for Result-Oriented Users by visually isolating http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7217921
[9] J. Liu, E. Pacitti, P. Valduriez, and M. Mattoso, “A
the relevant actions (submission and retrieval of Workflows), Survey of Data-Intensive Scientific Workflow Management,”
Second, it reduced the cost of building the Workflows for Journal of Grid Computing, mar 2015. [Online]. Available:
the Developer-Oriented users by providing re-usable Sub- http://link.springer.com/10.1007/s10723-015-9329-8
[10] N. Cerezo, “Conceptual workflows,” Ph.D. dissertation, Universite’
Workflows that could be combined in higher-level Meta- Nice Sophia Antipolis, 2013. [Online]. Available: https://tel.archives-
Workflows. Finally, it created Meta-Workflows that had an ouvertes.fr/tel-00942559/document
increased Utility but the same Cost to execute thus increasing [11] E. Deelman, D. Gannon, M. Shields, and I. Taylor,
“Workflows and e-Science: An overview of workflow system
the final Value. features and capabilities,” Future Generation Computer Systems,
vol. 25, no. 5, pp. 528–540, may 2009. [Online]. Available:
V. C ONCLUSIONS http://www.sciencedirect.com/science/article/pii/S0167739X08000861
This research is still its infancy and further examination [12] Y. Gil, E. Deelman, M. Ellisman, T. Fahringer, G. Fox, D. Gannon,
C. Goble, M. Livny, L. Moreau, and J. Myers, “Examining the
will be necessary to evaluate its worth. Firstly, the model has Challenges of Scientific Workflows,” dec 2007. [Online]. Available:
now been used to describe systems that are closely related http://eprints.soton.ac.uk/271187/1/computer07.pdf
with each other and developed by interconnected communities, [13] E. Elmroth, F. Hernández, and J. Tordsson, “Three fundamental dimen-
sions of scientific workflow interoperability: Model of computation,
so it may fail when used to describe solutions based on language, and execution environment,” Future Generation Computer
different philosophies such as the HubZero platform. Should Systems, vol. 26, no. 2, pp. 245–256, feb 2010. [Online]. Available:
the proposed model succeed in describing additional platforms, http://www.sciencedirect.com/science/article/pii/S0167739X09001174
[14] G. Terstyanszky, T. Kukla, T. Kiss, P. Kacsuk, A. Balasko, and Z. Farkas,
it could become the basis for further refinement and a more “Enabling scientific workflow sharing through coarse-grained interoper-
formal approach. ability,” Future Generation Computer Systems, vol. 37, pp. 46–59, 2014.
6
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
[15] K. Plankensteiner, R. Prodan, M. Janetschek, T. Fahringer, J. Montagnat,
D. Rogers, I. Harvey, I. Taylor, Á. Balaskó, and P. Kacsuk, “Fine-
Grain Interoperability of Scientific Workflows in Distributed Computing
Infrastructures,” Journal of Grid Computing, vol. 11, no. 3, pp. 429–455,
sep 2013. [Online]. Available: http://link.springer.com/10.1007/s10723-
013-9261-8
[16] National Institute of Standards and Technology, “Common
Industry Specification for UsabilityRequirements(CISU-R),”
http://zing.ncsl.nist.gov/iusr/documents/whatistheCISUR.html, 2006,
[Online; accessed 10-May-2016].
[17] ——, “Visualization and Usability Group,”
http://www.nist.gov/itl/iad/vug/index.cfm, 2016, [Online; accessed
10-May-2016].
[18] J. L. Gabbard, D. Hix, J. E. Swan II, M. A. Livingston, T. H. Höllerer,
S. J. Julier, D. Brown, and Y. Baillot, “Usability Engineering for
Complex Interactive Systems Development,” in Proceedings of Human
Systems Integration Symposium 2003, Engineering for Usability, jun
2003, p. 1.
[19] “SCI-BUS Project.” [Online]. Available: https://www.sci-bus.eu/
[20] “SHIWA Project.” [Online]. Available: http://www.shiwa-
workflow.eu/project
[21] “ER-FLOW Project.” [Online]. Available: http://www.erflow.eu/
[22] P. Kacsuk, Z. Farkas, M. Kozlovszky, G. Hermann, A. Balasko,
K. Karoczkai, and I. Marton, “WS-PGRADE/gUSE Generic DCI Gate-
way Framework for a Large Variety of User Communities,” Journal of
Grid Computing, vol. 10, no. 4, pp. 601–630, 2012.
[23] “CloudBroker Project.” [Online]. Available:
http://cloudbroker.com/platform/
[24] “DataAvenue Project.” [Online]. Available: https://data-avenue.eu/
[25] S. Herres-Pawlis, A. Hoffmann, T. Rosener, J. Kruger, R. Grunzke,
and S. Gesing, “Multi-layer Meta-metaworkflows for the Evaluation of
Solvent and Dispersion Effects in Transition Metal Systems Using the
MoSGrid Science Gateways,” in 2015 7th International Workshop on
Science Gateways. IEEE, jun 2015, pp. 47–52. [Online]. Available:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7217928
[26] G. Pierantoni and E. Carley, “Metaworkflows and Workflow
Interoperability for Heliophysics,” in 2014 6th International Workshop
on Science Gateways. IEEE, jun 2014, pp. 79–84. [Online]. Available:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=6882073
[27] G. Pierantoni, D. Perez-suarez, and P. T. Gallagher, “A WORKFLOW-
ORIENTED APPROACH TO PROPAGATION MODELS,” vol. 15,
no. 3, pp. 271–291, 2014.
[28] G. Pierantoni and C. Eoin, “HELIOGate: A Portal for HelioPhysics,” in
Science Gateways for Distributed Computing Infrastructures, P. Kacsuk,
Ed. Springer International Publishing, 2014, ch. HELIOGate,, pp.
195–207. [Online]. Available: http://link.springer.com/10.1007/978-3-
319-11268-8
7