We are seeing a massive growth in system deployments of infrastructure using elastic computing in the ‘as a service’ model. Modelling and Simulation as a Service (“MSaaS”), a concept following this trend, would deliver obvious benefits; these include flexibility of available training capability to meet surge needs and cost savings by reducing under-utilization of expensive simulation assets. MSaas has therefore been the focus of recent research sponsored by UK MOD.
Envitia has been involved in the Dstl Synthetic Environment Towers of Excellence (SE Tower) program since 2014. We have used our experience in information management and information exploitation gained in the geospatial domain to implement an experimental Synthetic Environment ecosystem using Envitia’s GeoRegistry. This has enabled the Modelling and Simulation domain to effectively discover and re-use of simulation assets.
Envitia has co-authored a technical paper with SEA, one of the partners in the SE Tower Architectures, Interoperability and Management of Simulations (AIMS) column, that will be presented at I/ITSEC 2017. I/ITSEC is the world’s largest modelling, simulation and training event, (their words, not mine) and attracts large numbers of exhibitors (in excess of 350) as well as defence, industry and educational personnel from across the globe.
Our paper “Information Management: A Core Enabler of the MSaaS Ecosystem” highlights the importance of understanding the information that you want to manage. The abstract is repeated below, and as with all technical abstracts it is a bit “dry”.
“The concept of Modelling and Simulation as a Service (MSaaS) has been a topic of investigation for the global modelling and simulation community in recent years, due to the potential it offers for the delivery of flexible, modular simulations in which services and resources can be re-used to deliver many different applications. Rapid reconfiguration of existing resources to provide new capability has value to many different simulation stakeholders, and service-oriented architectures offer a novel approach to achieving this goal.
In order to deliver the MSaaS vision, a move to a new architecture is required in which individual simulation services and resources exist independently and are used to compose a wide range of simulation systems. This type of architecture delivers an ecosystem. A core enabling technology for the implementation of an MSaaS ecosystem is a comprehensive and flexible information model underpinning all participation by services, resources and human actors. This information model must support and enable many different types of components in accordance with many different standards in order achieve the required levels of interoperability and re-use.
This paper presents a high-level, implementation-independent approach to information management for MSaaS, based on a developed multi-layered information model. This model contains an information layer, defining the information and data that exists within the ecosystem, a metadata layer, and a registry layer, which defines a structure for a searchable registry in which data and metadata, and their associations, are catalogued and managed. The paper then specifies how and why this type of approach to information management is a pre-requisite for achieving all of the benefits that MSaaS has to offer”
It may seem self-evident that you need to understand the information you want to manage, but analysis of the information types is often neglected. Without analysis it is impossible to understand the important features and characteristics of an asset, how the assets relate to each other and how as asset can be adequately described so that a user can evaluate it before use. This simple diagram sums up the basic approach that we took within the SE Tower work.
Key to success of the information management is the need to understand how the information will be used; remembering that different groups of people will use the same information in different ways so don’t be too self-focused when looking at the information models. In the diagram you can see that all of the information layers are informed by how they will be used.
The overall information model was developed in a platform neutral manner, but eventually it had to be implemented to prove it was an effective approach. The registry and metadata layers were successfully implemented using Envitia’s GeoRegistry software, accessible through open standards interfaces (ISO, OGC, OASIS etc.) to promote interoperability; one of the central tenets of the AIMS column. The use of open standards interfaces enabled other 3rd parties to create, store, describe, discover, evaluate and re-use simulation assets through manual and automated processes without recourse to proprietary software or standards.
All-in-all the concept of describing simulation assets in an open way and then effectively cataloguing these for re-use has proved to be highly successful in the prototype phase, though there is more work to do to exploit it operationally, particularly in areas of ontologies, access control and governance. But the work done does provide a strong indication of the tangible benefits of an MSaaS model as well as providing a practical example of how these benefits could be realised.
By Brendan Mason, Technical Lead Consultant, Applied Research Team