Page tree
Skip to end of metadata
Go to start of metadata

The attributes marked with a * are confidential and should not be disclosed outside the service provider.

Service overview
Service namegCube / D4Science DataMiner
Service areaData processing and analytics
Service phase

Production

Service description

This service offers a web-based workbench for data analytics compliant with Open Science practices. From the end user perspective, it offers a collaborative-oriented working environment where users:

  • can easily execute and monitor data analytics tasks by relying on a rich and open set of available methods either by using a dynamically generated web-based user- friendly GUI or by using a RESTful protocol based on the OGC WPS Standard;

  • can easily share & publish their analytics methods (e.g. implemented in R, Java, Python, etc.) to the workbench and make them exploitable by an automatically generated web- based GUI and by the OGC WPS protocol;

  • are provided with a “research object” describing every analytics task executed by the workbench enabling for repeatability, computational reproducibility, reuse, citation and provenance. The research object contains every input & output, an executable reference to the method as well as rich metadata including a PROV-O provenance record;

The data analytics framework is integrated with a shared workspace where the research objects resulting from the analytics tasks are automatically stored together with rich metadata. Objects in the workspace can be shared with coworkers as well as published by a catalogue with a license governing their uses. Moreover, the framework is conceived to operate in the context of one or more Virtual Research Environments, i.e. it is actually made available by a dedicated working environment offering (besides the framework and the workspace) additional services including those for managing users, creating communities, and supporting communication and collaboration among VRE members.

The data analytics framework is conceived to give access to two typologies of resource:
a distributed, open & heterogeneous computing infrastructure for the real execution of the analytics tasks. This distributed computing infrastructure is capable to exploit resources from the EGI infrastructure.

the pool of methods integrated in the platform, i.e. each method integrated in the framework is made available as-a- Service to other users according to the specific policy;

More details on this framework is available at https://wiki.gcube-system.org/gcube/DataMiner_Manager

Customer group

Any Research Performing Organization willing to provide its scientists with an Open Science compliant data analytics platform.
User group

The service is not tailored to serve the needs of a specific community. Rather, it is community agnostic and highly and easily customizable thus to serve the needs of a given community. Customization is achieved by configuring the instance serving a certain Virtual Research Environment with (a) the set of methods to be made available as-a-Service and (b) the resources forming the distributed computing infrastructure dedicated to execute the analytics tasks.

Up to now it has been and is successfully used by a quite rich array of diverse communities, namely those associated with the supported projects,, e.g., i-Marine (fisheries and marine biodiversity scientists), BlueBRIDGE (fisheries and aquaculture scientists, educators & SMEs), SoBigData.eu (social mining scientists), ENVRI+ (environmental scientists), AGINFRA+ (agriculture scientists), EGIP (geothermal scientists).

Value

The analytics platform is conceived to serve the needs of scientists (in particular, those belonging to the so called long-tail of science) by providing them with an easy to use working environment (nothing need to be installed on users’ machine). It is conceived to hide the technicalities related with the execution of tasks by relying on distributed computing infrastructures. Moreover, it is conceived to be exploitable by third-party software/applications, e.g. R-Studio, Q-GIS, or any workflow management system or application capable to interface with a RESTful service.

Worth highlighting that the platform is Open Science “compliant” (e.g., every method is “published” and citable, every task leads to a research object) and Virtual Research Environment friendly, i.e. it is conceived to be customizable with respect to the methods to offer in a given application context as well as it is conceived to benefit from a collaborative environment for sharing artefacts and comments.

Its characteristics make it particularly suitable to serve typical scientific contexts of the long tail of science.

TaglineOpen, user friendly and extensible data analytics platform ready for Open Science and VREs.
Features
Service options
OptionNameDescriptionAttributes
1
2



3



Access policiesPolicy-based and Wide-use. D4Science operates a number of instances of this 
Service management information
Service owner *D4Science.org
Contact (internal) *info@d4science.org
Contact (public)info@d4science.org
Request workflow *

Service request list

Ask to join one of the existing instances hosted by VRE, e.g. the ENVRIplus VRE https://services.d4science.org/group/envriplus/

For having dedicated instances, please contact info@d4science.org

Terms of use
SLA(s)
Other agreements
Support unithttp://support.d4science.org
User manual
Service architecture
Service components

#TypeNameDescriptionTRL [1]
1

Enabling


DataMiner Manager

DataMiner is a Web Service running in a Tomcat container equipped with gCube SmartGear. The Web Service exposes an interface compliant with the Web Processing Service (WPS) standard. DataMiner is able to exploit the heterogeneous resources offered by the D4Science e-Infrastructure to both retrieve and store data. The service allows users to execute community developed algorithms, written in several programming languages (e.g. Fortran, R, Java). Through the WPS standard, DataMiner provides three main access servlets:

GetCapabilities: Allows a client to request and receive service metadata (Capabilities), which describe the algorithms supplied by the service. Such operation provides the name and the general description of each process;

DescribeProcess: Allows a client to request and receive detailed information about the processes that can run on the service instances. Information includes the input and the expected output.

8 (at least)
2Enabling

DataMiner SmartExecutor

The DataMiner SmartExecutor executes tasks, i.e. functionally unconstrained pieces of code that lack a network interface but can be deployed into the service and executed through its interface. Example of tasks are R scripts and Java programs. An instance of SmartExecutor publishes descriptive information about the running environment. It can either execute tasks upon request or schedule the execution at configurable intervals on behalf of clients. Clients may interact with the Executor service through a Java library of high-level facilities that subsumes standard service stubs to simplify the discovery of available tasks in those instances. Each client can request to execute a Task or gathering information about the state of their execution. The client library simplifies the following activities: a) discover service instances that can execute the target task; b) launch the execution of the task with one the discovered instances; c) monitor the execution of the running task.

8 (at least)
3



Finances & resources
Payment model(s)Free
Pricing
Cost *
Revenue stream(s) *
Action required



[1] Technology Readiness Levels (TRL) are a method of estimating technology maturity of components during the acquisition process. For non-technical components, you can specify “n/a”. For technical components, you can select them based on the following definition from the EC:

  • TRL 1 – basic principles observed
  • TRL 2 – technology concept formulated
  • TRL 3 – experimental proof of concept
  • TRL 4 – technology validated in lab
  • TRL 5 – technology validated in relevant environment (industrially relevant environment in the case of key enabling technologies)
  • TRL 6 – technology demonstrated in relevant environment (industrially relevant environment in the case of key enabling technologies)
  • TRL 7 – system prototype demonstration in operational environment
  • TRL 8 – system complete and qualified
  • TRL 9 – actual system proven in operational environment (competitive manufacturing in the case of key enabling technologies)

1528
1530