Workshops‎ > ‎cs-ga-2012‎ > ‎

Andrew Hutchison and Roland Rieke


Biography:

Andrew Hutchinson


Dr Andrew Hutchison is part of the International Presales Management group of T-Systems International, having moved into this role after establishing the internal Process, Quality and IT (PQIT) group activities of the South African business unit of T-Systems. For the international organization he currently focuses on the Presales groups and activities in the USA, Canada, Mexico, Brazil, UK, Switzerland & South Africa. Andrew was previously General Manager of the Telecommunication Services division in South Africa. He has also held executive responsibility for business development in South Africa, as well as for internal IT and corporate Security portfolios.

 

Prior to joining T-Systems, Andrew was co-founder and director of an IT consulting and training company, Computer Science Institute, which was acquired by the Software Futures Group. Andrew has previously also been a full time member of the academic staff of the Computer Science Department at the University of Cape Town (UCT) and has a PhD in Computer Science from the University of Zurich, which he completed while working at the IBM Zurich Research Laboratory in Switzerland. He has been a visiting scientist at the University of Southern California in Los Angeles, and is currently appointed as an Adjunct Professor in the UCT Computer Science Department from 2009 to 2014.  



Roland Rieke



Roland Rieke works since 1982 as a senior researcher at the Fraunhofer Institute for Secure Information Technology SIT. His research interests are focused on the development of methods and tools for formal security models and application of these techniques for architecting secure and dependable systems. In the project EVITA (E-safety Vehicle Intrusion proTected Applications), for instance, he worked on a method for security requirements elicitation in systems of systems applied in the context of vehicular communication systems. He is currently working on predictive security analysis for event-driven processes in the context of the Internet of things within the project ADiWa (Alliance Digital Product Flow). His recent papers furthermore comprise work on attack graph analysis and on construction principles for dependable and secure parameterised systems. Roland is the research director of the project MASSIF (MAnagement of Security information and events in Service InFrastructures), a large-scale integrating project co-funded by the European Commission. He is member of the strategy board of the Effects+ (European Framework for Future Internet Compliance, Trust, Security and Privacy through effective clustering) project and member of the ERCIM working group on Security and Trust Management.





Abstract:
Measuring Progress in Cyber-Security: An Open Architecture for
Security Measurement Consolidation

It is sometimes said that “what you cannot measure, you cannot manage”. Cyber Security is an area of great global focus, yet it is both hard to manage and - arguably - even harder to measure. But the two concepts go together: if some sort of measurement approach could be implemented, it should at least be possible to assess whether progress (or regress?) is being made in the difficult area of Cyber Security. In this paper, we present some ideas on global measurement of cyber security, but propose an approach whereby such activity can be started in a “bottom up” fashion so that at least at enterprise or organisational level some steps towards “measured” security can be applied. In the future this could possibly be extended to a national or international scale.

In spite of the fact that technical security solutions are deployed, there are numerous instances of processes or transactions being compromised. Without a holistic view from business process to logical and technical security realisation, there is high potential for gaps or mismatches to occur. Fueling this situation is the fact that life-cycle approaches to security are not easily applied -- or measured. In this paper we argue for a meta-model approach to drive security from purpose to practicality, through an analysis and refinement approach, and also for a security measurement approach in support of local, enterprise, regional, national and even international consolidation of security health information. 

To achieve a meta-model approach for security, we present a Security Information Meta Model consisting of:

·        high level goal setting,

·        security requirements,

·        measurement requirements, and

·        objects of measurement.

Through applying this model, high level goals for security can be established and defined. Importantly measurement objectives are also developed and stated at this point. The ISO27004 standard outlines what these objectives could be, but we believe that they are hard to incorporate unless done as (and when) we are proposing.

By proceeding in this way, security is designed in a manner which can be tested against. Necessarily, activities of analysis and refinement are required to move from security requirements to measurement requirements. In particular objects of measurement also need to be identified.

If the security semantics can be defined in this way, an interesting “control” model can be applied to systems based on a so-called on/if/do/why approach. What is meant by this?

On the occurrence of a specific event, then if a particular condition occurs the system will do an action and assess why in terms of security pertinence.

As an example of an on/if/do/why approach, a core information system which is integrated in a Security Information and Event Management (SIEM) environment could be constructed along these lines. By providing an event correlation engine as one component, event conditions can be identified to provide the on element. Similarly application and/or network state (and possibly predicted state) can be interrogated to provide the if element. Depending on how the condition evaluates, the do element of  decision support and/or policy decision making can be applied. A security information modeller can assist in the why determination. 

Through such an approach a system can be measured and, arguably, better managed. By starting to apply this approach on a micro-scale, it can ultimately be scaled and expanded to allow for similar construction for enterprise, national or even international systems.

In this way different facilities can be integrated to provide a “measured” security view.

To achieve the measurement objective, the concept is to define measures very explicitly for a system -- and to be able to measure the security effectiveness against such measures. By aggregating systems composite measurement can be performed.

An area under consideration is whether measurement attributes could be incorporated into an SIEM platform like MASSIF (http://www.massif-project.eu ) (a large-scale integrating project co-funded by the European Commission), to enable ongoing assessment of security effectiveness -helping practitioners to “manage”, based on a rich set of “measurements” if these can be built into the system.


Comments