Jump to content Jump to navigation

Information infrastructure

In the Netherlands and far beyond, Naturalis occupies a unique position due to its combination of strong research, a gigantic collection and a heavily visited museum.

This combination also gives to an exceptional situation in the area of ICT. Not only must the standard office automation work well but sometimes very specialized requirements in these three areas also need to be met.

Science, collection, museum

For example, from the scientific perspective high requirements are placed on the performance of systems and use is made of specialized research software. The collection and its digitization require a good structuring and disclosure of large quantities of data through web services. And the museum, which will be completely refurbished over the next few years, poses requirements regarding the presentation of relevant information about biodiversity and how visitors interact with this.

However, a lot of overlap exists between these areas as well. Both the museum and collection require the ICT services to always work and be available. And all systems have development pathways in which code is developed, applications are tested and systems are made operational.

So to serve all three specializations efficiently and effectively not just the differences but also the similarities between the specializations need to be taken into account.

Making use of similarities

The biggest similarity between the three areas is the aforementioned development process. Although different types of applications and services are developed, the pathways these follow are broadly speaking the same. Furthermore, interesting developments in ICT are taking place in this particular area.

One of the developments is DevOps, where ways are sought from organizational and software development perspectives to better connect the different steps in the development pathway and to improve the collaboration between developers, testers, and system managers.

Within Naturalis efforts are constantly being made to apply this philosophy to the three development areas. For example OpenStack is being used as a cloud platform for testing and deploying systems. Puppet is being used for configuration management so that the reproduction and consequently testing and deployment of systems is made easier and more reliable. Finally Github is being used by researchers, developers, and managers for developing and sharing code.

For architecture the differences initially appear to be greater. Of course the solutions required differ per pillar. However, by dividing infrastructure up into different layers, similarities in this area also become clear.

The ‘lowest’ hardware layer has the biggest similarity between collection and science, although in the case of science more powerful servers are required. The hardware required for the museum differs from this because direct interaction with visitors is required. This is realized through the use of small form factor PCs and the open hardware platform Arduino.

For the Infrastructure as a Service (IaaS) layer above this, the science and collection system are hosted on the basis of OpenStack. Furthermore, dependent on the developments it is also being examined whether control of hardware in the new museum can be realized in a similar manner based on OpenStack or a related technology.

The Platform as a Service (PaaS) layer is formed by offering scientific development environments based on Python and R, for example. In the case of collection-related services, the Netherlands Biodiversity API forms a key component that is used through the standard Drupal development platform. The development of museum applications, which will communicate with a museum message bus, based on HTML5 web technology and Processing, will be facilitated in a similar manner. Monitoring of these systems will take place using Elasticsearch, Logstash and Kibana.

Finally, in the topmost Software as a Service (SaaS) layer, are the applications offered to researchers, museum visitors, and interested professional and amateur biologists. Research applications such as Galaxy, the internally developed Bioportal, and the museum exhibit that can soon be used in the refurbished museum, are based on the same reusable web technology. This level can therefore effectively meet the diverse range of needs.