Ecosystem for model-driven security is getting ready

Many vendors provide model-driven tools today, incl. business process management (BPM), model-driven engineering/development (MDE/MDD), model-driven integration (MDI), enterprise architecture (EA) etc.
Process-let SOA orchestration and model-driven code generation or service integration is also a reality today, and big vendors such as Microsoft have announced that they will release these features in their mass-market software development tools.
This is great news for model-driven security, which ties into model-driven tools in order to automatically and traceably produce fine-grained, contextual security policies.
The fact that mainstream tools are available and in use today enables shrink-wrapped, push-of-a-button model-driven security to be added to such model-driven tools - ObjectSecurity has just produced such a shrink-wrapped security policy generator for IntalioBPMS and their OpenPMF model-driven security technology.
SOA security, and specifically security policy management for SOA are also being closely examined, and model-driven security has been identified as a great solution.
So everything is finally coming together in the mass market - watch this space!

Revisited: Aligning business and IT security

Well, ok, there are a number of useful best practice guidelines for information security management, e.g. ISO 27000 family, COBIT, ISMS, ITIL, which help communicate the processes and requirements to management, select controls, and measure success. So far, so good. However, these documents are so business-centric that IT security enforcement does not simply "fall out at the bottom".
But beware: What ultimately matters is not the documentation produced, but the actual enforcement across your information systems. Real-world attacks are obviously not thwarted with documents!
Today, mapping the requirements from the produced documents down to concrete IT enforcement (and bringing measurements back up) is typically done in a pretty ad-hoc way. The focus is primarily on what the management wants to (not) see, rather than in what matters.
Model-driven security (e.g. OpenPMF), tied into an overall security management methodology, can help close (parts of) this gap in a traceable, runtime, automatic way - therefore model-driven security should be a critical element of effective "business-driven compliance management".

Why "business process-led" model-driven security is useful

You may have seen that ObjectSecurity released a beta for OpenPMF with Intalio BPMS that supports "business process-led" model-driven security (for BPMN). This feature will form part of the new OpenPMF 3.0 release planned for Q1/2009. Why are business processes so relevant for security policies? Simply because the workflow context is a powerful contextual element of a fine-grained security policy. For example, an e-shop can block access to their bank's credit card charging facility for any access except at the specific step in the business process workflow when the access should be granted. Control is therefore much more fine-grained and contextual than for example role-based or label-based access control. Research (e.g. here) has focussed around the addition of security to business processes for a while, and this feature is now implemented (in beta) in OpenPMF for Intalio's open source BPMS.

Model-driven security needs to be cross-platform

Another observation we made over the last couple of years is that there will most likely be no "one size fits all" technology platform (e.g. middleware) in today's large, complex IT environments. A plethora of platforms (e.g. web services, JMS, CORBA, CCM, DDS) will probably be used, potentially orchestrated using some BPM technology (e.g. BPMN/BPEL) or model-driven integration (MDI) technology.
As a result, model-driven security needs to be able to ensure correct policy generation and enforcement for all these platforms.
OpenPMF supports policy enforcement for a large number of enforcement points, including web services, JMS, CORBA, CCM, DDS. XACML is also supported to ensure the emerging SOA enforcement landscape can be supported.
Push-button policy generation using model-driven security from a single place only if enforcement is supported cross-platform.

*New publications* about model driven security

ObjectSecurity published a paper at ISSE 2008 with a concrete model-driven security healthcare example where a HIPAA healthcare compliance requirement is mapped to cross-platform IT infrastructures including BPM, web services, and CCM. The presentation is here, and the detailed paper is published at:

Lang U., Schreiner R., "Managing business compliance using model-driven security management", in Pohlmann N., Reimer H., Scheiner W. (editors), Proceeedings ISSE 2008 Securing Electronic Business Processes - Highlights of the Information Security Solutions Europe 2008 Conference, Vieweg + Teubner, ISBN 978-3-83480660-4, Edition 2009

Abstract: Compliance with regulatory and governance standards is rapidly becoming one of the hot topics of information security today. This is because, especially with regulatory compliance, both business and government have to expect large financial and reputational losses if compliance cannot be ensured and demonstrated. One major difficulty of implementing such regulations is caused the fact that they are captured at a high level of abstraction that is business-centric and not IT centric. This means that the abstract intent needs to be translated in a trustworthy, traceable way into compliance and security policies that the IT security infrastructure can enforce. Carrying out this mapping process manually is time consuming, maintenance-intensive, costly, and error-prone. Compliance monitoring is also critical in order to be able to demonstrate compliance at any given point in time. The problem is further complicated because of the need for business-driven IT agility, where IT policies and enforcement can change frequently, e.g. Business Process Modelling (BPM) driven Service Oriented Architecture (SOA). Model Driven Security (MDS) is an innovative technology approach that can solve these problems as an extension of identity and access management (IAM) and authorization management (also called entitlement management). In this paper we will illustrate the theory behind Model Driven Security for compliance, provide an improved and extended architecture, as well as a case study in the healthcare industry using our OpenPMF 2.0 technology.

We are pleased that the papers from the MODSEC 2008 (Modeling Security Workshop) are now also available online here (CEUR Workshop Proceedings),

Please contact us if you have any products or publications you would like to see covered in this blog.

Model Driven Security & SOA - take the survey & get involved

The UK Cyber Security KTN currently runs a SOA security analysis project (see www.secure-soa.info) and there is a study about the main concerns of SOA security by end-users.
Please click here www.secure-soa.info to take the 5 minute survey, and get involved in the email group, wiki, and report!

Business-driven security: Aligning business and IT security

Organizations today advocate the architectural vision of "aligned business and IT" - this means that the organization's IT landscape is aligned with achieving business goals, and that it can be adapted quickly to help the business respond to changes (e.g. in the market). Enterprise Architecture (EA), Business Process Management (BPM), BPM-driven Service Oriented Architecture (SOA), Model Driven Integration (MDI), and Model Driven Engineering (MDE/MDA) are examples that tie into such a vision.

Model-driven security is a critical aspect ofthis vision because - in line with the overall vision - it allows 1) business security requirements to be defined, 2) these requirements automatically transformed into IT-centric security rules, 3) automatically enforce the rules across the IT landscape, and 4) demonstrate compliance to the business.

The result is a closed loop from the business to IT and back to the business. The benefits include: enable IT/business agility, save cost, improve security, and of course align business and IT security.

Analyst firms forecast the mainstream for model-driven, process-led approaches within 5 years, and model-driven security is set to piggyback onto that adoption. So it is time to look into it now. Feel free to read our white paper at http://www.openpmf.com/.

"Security stove-piping" & agility (e.g. SOA)

It is clear that end-users are trying to get away from stove-piped, hard-coded IT environments. Instead, they want agile, reconfigurable, modular IT environments, as e.g. advocated by Service Oriented Architecture (SOA). A lot of effort has been put into architecting modular, model-driven approaches to achieve system agility.

Unfortunately security typically gets overlooked, and traditional security tools are deployed and configured (e.g. manually configured policies set in app servers, IAM systems etc.). The result is a system that is almost as stove-piped as before. ObjectSecurity calls this problem "security stove-piping".

Model driven security as a security management approach enables agility and security, and is therefore a critical ingredient in the SOA security mix. Contact ObjectSecurity if you would like to discuss this further.

Management vs. interoperability: Model driven security vs. today's authorization management

It is clear that the number of fine-grained IT authorization policies that are spread across a medium-size or large-size IT environment can easily go into the 10,000's and 100,000's. Just take the rules from firewalls, databases, and single sign-on systems, and you see that the complexity has grown out of hand: Security is simply unmanageable.

Today's authorization management solutions (sometimes called "entitlement management") tackle the problem by simply putting all the complexity into a single place (the Policy Access Point, PAP). By and large the rules in the central manager are still at the same semantic level and complexity as the rules that are spread across the IT environment if no authorization management is used. This is clearly not a significant reduction of complexity.
(By the way, identity management does not actually cover this problem very well, as it is pretty much concerned with managing identities and less with the management of fine-grained, expressive, maybe context-sensitive authorization policies).
In summary, today's authorization management makes the problem evident, rather than solving it.

What today's vendors are good at is solving the policy interoperability challenge: XACML is a webservice standard for exchanging authorization policy information, and vendors include ObjectSecurity, Cisco, CA, etc.

Model driven security is concerned with solving the complexity challenge: It lets you manage simple, business-driven security policies, and generates the 100,000's of rules for the particular deployment automatically. Sounds like magic, but it is not. Contact ObjectSecurity, the leading model driven security vendor if you would like to learn more.

So in summary: authorization management is necessary but not sufficient.

Model driven security recognized as impactful, innovative, intriguing by leading analyst firm

Model driven security has recently received internationational recognition indicating the continued emergence of the approach: ObjectSecurity has been named “Cool Vendor” by Gartner, Inc, the leading IT analyst firm, for our innovative OpenPMF 2.0 technology. OpenPMF 2.0 is the first full “Model Driven Security Management” product in the market. Model driven security is predicted to be a high-impact technology area that helps deal with the complexity of today’s authorization management technologies. Read the press release and some information about OpenPMF 2.0 here:

Press Release ObjectSecurity Named "Cool Vendor" by Leading Analyst Firm
(Cambridge/UK – 04 April 2008) – ObjectSecurity, the leading solutions provider for Model Driven Security Management and secure information sharing in mission-critical industries such as air traffic control, today announced that Gartner, Inc., the world's leading information technology research and advisory company, has named ObjectSecurity in its "Cool Vendors in Application Security and Authentication, 2008”. The April 04, 2008 report was written by Ray Wagner, Joseph Feiman, Neil MacDonald, Arabella Hallawell, Ant Allan, and Gregg Kreizman. According to the report, vendors selected for the "Cool Vendor Report" are innovative, impactful and intriguing.
"We are honored to be included, which we believe is recognition by the world's leading information technology research and advisory company, Gartner," said Dr. Ulrich Lang, CEO and co-founder of ObjectSecurity.
About OpenPMF 2.0 - OpenPMF 2.0's powerful, yet easy-to-use technology is the only 'model driven security management' solution in the market today. It is the most flexible, extensible, standards based, and easy-to-use enterprise security management framework on the market. The patent-pending technology is based on 9 years of solid research and development by leading experts who are currently driving international standardization of model driven security. OpenPMF 2.0 is the most thought-through solution on the market and listed as a promising high-impact technology on Gartner’s “Hype Cycle for Information Security 2007”. OpenPMF 2.0 benefits include reduced cost, improved enterprise-wide security compliance, and low-maintenance security management for agile Service Oriented Architecture (SOA). OpenPMF 2.0 lets you manage security at a business-driven, intuitive high level of abstraction close to human thinking. OpenPMF 2.0 is fully customizable so that you can define customized policies in the way you think about security in the context of your organization.
About Gartner's Cool Vendors Selection Process - Gartner's listing does not constitute an exhaustive list of vendors in any given technology area, but rather is designed to highlight interesting, new and innovative vendors, products and services. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness of a particular purpose. Gartner defines a cool vendor as a company that offers technologies or solutions that are: Innovative, enable users to do things they couldn't do before; Impactful, have, or will have, business impact (not just technology for the sake of technology); Intriguing, have caught Gartner's interest or curiosity in approximately the past six months.
ABOUT OBJECTSECURITY - ObjectSecurity Ltd. is a Cambridge (United Kingdom) and San Jose (CA, USA) based world-leader in model driven security and authorization management. The company offers the ground-breaking model-driven OpenPMF enterprise security management ecosystem and various secure middleware platforms. ObjectSecurity provides services for model driven security, middleware security, SOA security, secure information sharing (e.g. CDM). Their customer base includes Agilent Technologies, BAA Heathrow Airport, Deutsche Telekom, ESG, European General Electric, Intel, QinetiQ, Royal Bank of Scotland, Real-Time Innovations, Twinsoft/Hewlett-Packard, US Naval Research Lab and others. ObjectSecurity specializes on information security for complex IT environments in mission-critical markets.
PR CONTACT Dr. Ulrich Lang, ObjectSecurity Ltd., info@objectsecurity.com, www.objectsecurity.com, +44 1223 420252/+1-800-898-9148
>>> PDF version: http://www.objectsecurity.com/doc/20080407_gartnercoolvendor.pdf
>>> Purchase the report: http://www.gartner.com/7_search/Search2Frame.jsp?keywords=objectsecurity

The need for standards *NOW*

We need standards for Model Driven Security. There are primarily two reasons for this:

1. we need to avoid vendor lock-ins, because they will hurt end-users and vendors alike.
2. we need to avoid market fragmentation into dozens of products that have their own way of expressing security models

If industry is not commited to preventing vendor lock-ins and market fragmentation, then Model Driven Security would take much longer to become mainstream. Also, the shakeout in the market would be bloodshed, where innovation typically goes out of the window.

As a consequence, ObjectSecurity and several OMG members have come together at the Object Management Group (OMG) consortium to work towards a Model Driven Security Policy standard. This standard should specify a common vocabulary which allows policies to be transferable between different vendors' tools. An RFI has just been issued by the OMG.

Please contact us if you would like to know more about this.

Model Driven Security, accreditation, and agile SOA

Defence companies and DoD/MOD are increasingly aware that they are "hitting the wall running" with SOA and certification/accreditation (e.g. Common Criteria).

The main problem is simple: SOA is about agility, and dynamically responding to change by allowing fast reconfiguration of the infrastructure "Lego blocks". Accreditation is about accrediting the assurance of a static system using some elaborative analysis process.

Unless the two are brought together, there will simply be no useful SOA in defence. Model Driven Security (MDS) can help achieve this.

IBM has recently (6/2007) published a Working Paper on the subject. It is not very dense, essentially they are saying that the challenges are due to complexity. The relevant information includes:
1) they then say "the new direction parallels the way Model Driven Architecture (MDA) and Model driven Development (MDD) have restructured the ... challenges and have provided architects ... better leverage over SOA complexity".
2) cultural and the accreditation community needs to be trained
3) support incremental change in accreditation practices
4) SOA should be deployed without agility (why would you buy SOA then at all?)
5) traditional accreditation approaches need to be adapted to match SOA better
6) security mechanisms are there and aren't really the problem

The first point echoes what ObjectSecurity has said since 2005: Model Driven Security is a highly useful concept to provide accreditable, agile SOAs with low-maintenance security policy management.

Please contact us if you would like to know more about agile SOA security and accreditation.

"Security Stove-Piping" and Model Driven Security

It turns out that one of the main security issues related to SOA is that security is typically implemented in such a way that it cannot preserve the agility SOA (without security) promises. The reason behind this is simple: If I have a large SOA with many interactions, and I reconfigure (e.g. orchestrate) the SOA, I will need to check all the security policies and figure out whether anything changed. It is likely that a significant SOA reconfiguration changes security policies for many nodes.
Now imagine having to do such a manual process everytime you reconfigure the SOA - clearly not cost-effective and highly error-prone.
We at ObjectSecurity call this "security stove-piping".
Model driven security (as implemented in the patent-pending ObjectSecurity's OpenPMF 2.0) allows you to state your security intent in an intuitive, general, and undistorted way that remains relatively constant over time.
The semantic gap between this high-level intent and what needs to be enforced on the SOA infrastructure layer is then bridged using model driven security. The concept is related to Model Driven Architecture (MDA), and applied to security e.g. in our OpenPMF 2.0 SecureMDA sub-module.
The benefits are intuitive: As long as my high-level intent remains the same, I can reconfigure the SOA without any changes to the abstract security policy models. Contact us if you would like to know more about how this works in OpenPMF 2.0's TrustedSOA submodule.

By the way, if you happen to be in the area, then please feel free to sign up to our Peer2Peer session at the RSA Conference 2008, San Francisco, April 2008:

ObjectSecurity will present a peer-to-peer session "How can we secure SOA without losing agility?" at the RSA Conference 2008, San Francisco, CA, USA, 7-11 April 2008. Contact us to arrange a meeting.Abstract: In this Ask the Moderator session, ObjectSecurity discusses how SOA security must go beyond web services security. The core issue is how to specify and maintain consistent/effective security policies for *agile* SOA. This cannot be done manually (too complex/labor-intensive). New approaches such as Model Driven Security are needed. Session topics incl. security stove-piping, how to reduce cost/effort, architecture approaches, experiences, secure BPM SOA." (P2P-205A, 9 Apr 2008, 1:40 PM - 2:30 PM).

See you there!

Publications & Resources about Model-Driven Security

This blog also tries to provide a forum for publications about model driven security. Please put any abstracts into the comments of this message and we will merge them into the main message.

ObjectSecurity released a publication Model driven security for agile SOA-style environments, by Dr. Ulrich Lang & Rudolf Schreiner at ISSE 2007:

There is evidence that many IT security vulnerabilities are caused by incorrect security policies and configurations (i.e. human errors) rather than by inherent weaknesses in the attacked IT systems. Security administrators need to have an in-depth understanding of the security features and vulnerabilities of a multitude of ever-changing and different IT "silos". Moreover, in complex, large, networked IT environments such policies quickly become confusing and error-prone because administrators cannot specify and maintain the correct policy anymore. Agile service oriented architecture (SOA) style environments further complicate this scenario for a number of reasons, including: security policies may need to be reconfigured whenever the IT infrastructure gets re-orchestrated; security at the business process management layer is at a different semantic level than in the infrastructure; semantic mappings between the layers and well-adopted standardised notations are not available. This paper explores how the concepts of security policy management at a high, more intuitive (graphical) level of abstraction and model-driven security (tied in with model driven software engineering) can be used for more effective and simplified security management/enforcement for the agile IT environments that organisations are faced with today. In this paper, we illustrate in SecureMDA™ how model driven security can be applied to automatically generate security policies from abstract models. Using this approach, human errors are minimised and policy updates can be automatically generated whenever the underlying infrastructure gets re-orchestrated, updated etc. The generated security policies are consistent across the entire distributed environment using the OpenPMF policy management framework. This approach is better than having administrators go from IT system to IT system and change policies for many reasons (including security, cost, effort, error-proneness, and consistency). The paper also outlines why meta-modelling and a flexible enforcement plug-in model are useful concepts for security model flexibility.

---

Gartner released a study Model-Driven Security: Enabling a Real-Time, Adaptive Security Infrastructure that defines:

"Model-driven security is the use of visual models or domain specific modelling languages during application design, development and composition to represent and assign security primitives — such as confidentiality, integrity, authentication, authorisation and auditing — to application, process and information flows independent of the specific security enforcement mechanisms used at runtime."

---

ObjectSecurity released a study Model Driven Security - A new security management approach applied to SOA - please contact to puchase.

---

DEFINITION: MODEL DRIVEN SECURITY

ObjectSecurity is just finishing a larger study about model-driven security (MDS), as part of which they did an exhaustive search of different approaches, architectures, and definitions. In the end they settled for the following definition:

Model driven security (MDS) is the tool supported process of modelling security requirements at a high level of abstraction, and using other information sources available about the system (produced by other stakeholders). These inputs, which are expressed in Domain Specific Languages (DSL), are then transformed into enforceable security rules with as little human intervention as possible. MDS explicitly also includes the run-time security management (e.g. entitlements/authorisations), i.e. run-time enforcement of the policy on the protected IT systems, dynamic policy updates and the monitoring of policy violations.

Please put any suggestions into the comments field and we will modify this definition as needed.

As part of their study, they also analysed the product/vendor landscape in technical depth, and identified industry trends - this information can be made available upon request. Contact us here if you are interested in details or would like to purchase a report.

Gartner Hype Cycle for Information Security 2007

Gartner has just released their new Hype Cycle for Information Security 2007, and model driven security is on it. ObjectSecurity's OpenPMF 2.0 (www.openpmf.com) has been identified as aleading product in this emerging area.

This shows that Gartner believes that model driven security is a critical technology approach to simplify enterprise security.

This blog is a public forum and we are welcoming any views on this.

Related blogs

There is a related blog at www.trustedsoa.org and one with more of a middleware security focus at www.securemiddleware.org.

Looking for OpenPMF, SecureMDA, TrustedSOA?

Did you get to this page because you are looking for ObjectSecurity's model driven security tools to automatically generate security policies from models? If so, please go to:

http://www.openpmf.com
http://www.trustedsoa.com
http://www.securemda.com/

Model driven architecture and SOA assurance

Model driven engineering and SOA seem to complement each other very well. SOA enables horizontal decoupling of services (the issues of that are discussed on our other blog www.trustedsoa.org), and MDA enables horizontal decoupling (of models from implementations). Now this is obviously an oversimplification, but a nice upcoming architectural idea.
Security plays an important role here, and it is currently still a bit unclear to many how security can be defined and enforced in a manageable way. Of course there are webservices security specifications, but those (at least the ones that work in real-world products today) only deal with the protocol layer, which is the easy bit.
The harder bit is how to define and enforce policies for agile SOA-style enviroments. We at ObjectSecurity believe that model driven security (MDS) can help here because it allows to generate security policies for agile systems from a stable model.
But securing SOA is only one application of this useful concept...

Welcome & Introduction

On this blog we will discuss model driven security. Please feel free to comment.

Defining security policies for complex, large IT environments is a difficult, cumbersome, and error-prone task. This is in particular the case for agile IT environments such as highly distributed component based systems and Service Oriented Architecture (SOA). We have shown that model-driven security, which allows the generation of security policies from the application models, helps build and maintain secure, agile IT environments.

Today software modelling is the accepted best-practice approach for developing flexible and reusable software applications where abstract application models are turned into software using a modelling toolchain. The OMG Model Driven Architecture (MDA) is the leading standard framework for software modelling. The ObjectSecurity/Fraunhofer FOKUS SecureMiddleware includes a full MDA development toolchain.
Why not apply the same logic to security and automatically generate security policies and high assurance from the application models? This way, you can be confident that the deployed system matches the models, and that you have not forgotten any security policy aspects.
And most importantly, you can reconfigure and redeploy your (possibly distributed) applications by simple changes in the model - the underlying software and security policies will be automatically matched to your models through the automatic MDA and SecureMDA tool chains.

This approachhas been showcased by ObjectSecurity (with their SecureMiddleware partner Fraunhofer FOKUS) in their SecureMDA technology.

Any comments on model driven security are greatly appreciated.