As chief of the National Security and Intelligence, Surveillance and Reconnaissance Division for DSTO, Dr Tony Lindsay is well placed to see the country’s threat landscape and provide insights into the risks, adversaries and defences we need to consider.
At the recent Technology in Government forum held in Canberra he spoke about information integration and the security aspects of enterprise architecture, and the importance of policy.
One of the challenges, says Lindsay, isn’t a lack of information on threats and adversaries but getting access to it in a usable form. He noted that during the first Gulf War, his division spent 80% of their time gathering data and 20% doing analysis.
So what’s the answer?
“For us, part of the solution is having some sort of extensible architecture,” says Lindsay. “That includes the security architecture – the mark-ups, the metadata tagging for security visibility because those things will change very quickly”.
This is the important one - interoperability standards for enabling appropriate discovery. Having people spending 80% of their time just finding data is crazy’” he says.
The use of automation is critical says Lindsay. With the volume of data that needs to be analysed increasing and the continuing dearth of highly skilled personnel, significant automation will need to be employed in order to move Lindsay’s 80/20 observation.
Although Lindsay’s focus would seem to be on external security threats, he noted that internal policies and procedures are critical.
“Are people looking at the data they’re allowed to look at”? he asks.
In his work, Lindsay works with representatives from many allied governments around the world. Lindsay spoke about the US Distributed Common Ground System. This set of protocols describes information sharing procedures for US armed forces. When this system was developed, industry involvement at all levels was a critical factor in its success.
However, one of the lessons learned was that each section of the military used different contractors and systems resulting in interoperability issues. Lindsay notes the importance of not giving up responsibility for critical interfaces between systems as well as operational expertise and intellectual property so systems can be refined and augmented over time.
When looking at security systems, Lindsay noted there had been a tendency in the past to purchase an appliance or application and consider the issue to be managed However, with the threat and security landscape changing so quickly, there’s a need to be constantly vigilant and aware of changing risks and new technologies.
One of the initiatives Lindsay discussed was the Common Foundation Framework. This framework doesn’t mandate the use of specific applications or technologies but ensures that common design and security principles are adopted across different branches of government. This extends to a shared development environment so that there was a place where departments could test out information sharing safely to ensure interoperability was not compromised.Read more: Machine learning key to building a proactive security response: Splunk
Lindsay also discussed a “test bed” called ELIIXAR. It spans the Australian defence organisation. It is open source software so that vendor lock-in was avoided. It’s built on a scalable architecture so that it can be adapted as needed for different scenarios.
Importantly, ELIIXAR isn’t limited to Australian agencies. It can also operate with the United States and NATO for information sharing – an important consideration as multi-national forces often need to cooperate in many conflicts and other actions.
ELIIXAR uses off-the-shelf infrastructure with everything being browser based so bespoke applications aren’t needed. This delivers significant flexibility for defence agencies. Although significant progress has been made in DSTO Lindsay and his team aren’t resting on their laurels. He sees potential for many improvements.
One of these is identifying the priority mission threats – determining which mission threats are most critical with data coming from multiple source. Again, this harks back to his initial premise that a lot of time is spent on sorting through data rather than real analysis.
Another challenge was getting different government departments to identify metadata standards to aid exchange and interpretation within government.
Want to know more?
Why not become a CSO member and subscribe to CSO's mailing list.
Get newsletters, updates, events and more right here