KitPlus - The TV-Bay Magazine

To view this page ensure that Adobe Flash Player version 11.1.0 or greater is installed.

TECHNOLOGY Bruce Devlin Mr MXF Ltd. @MrMXF CLASS: Content security in the cloud Mr MXF thinks... Where did I leave my keys? Security is a difficult topic. There is no such thing as generic security. You always secure yourself against specific threats and hope the solution is generic enough to cover other similar classes of threat. For example – placing an expensive lock on the front door of your house prevents the specific threat of gaining access by using the door handle on the door. It gives you no protection at all from someone using a diamond tipped chainsaw to cut through the house to make a new door. (Yes this is a real thing – a friend of mine trains law enforcement officers to do this). One of the great opportunities that comes with cloud processing and cloud storage of content is that it’s new and it’s flexible and it allows new ways of collaborative working that cannot be done any other way. One of the commercial risks that need to be addressed with cloud processing is that of security. In any cloud deployment there will be multiple organisations involved. You may be buying Software As A Service (SaaS) to store, edit and manage your content. The SaaS provider is running parts of their software on top of another company’s Platform As A Service (PaaS) to provide some global service to their software (e.g. Authentication and Identity services) and that platform may be physically running on multiple Infrastructure As A Service (IaaS) datacentre instances to make the whole thing work. If security is a key commercial driver for you adoption of any cloud technology then understanding the threats to the different stakeholders in the chain is vital to the overall security model. Having great user authentication at the SaaS layer is good, but it won’t protect against someone with a diamond tipped software chainsaw gaining entry at the PaaS layer. Likewise the PaaS layer is only secure when the physical, network and software security at the IaaS remain intact and no-one leaves a back door open. 36 | KITPLUS - THE TV-BAY MAGAZINE: ISSUE 114 JUNE 2016 Cloud security is a fluid and ever changing topic with Best Practise being re-written with every new data breach. It’s important to realise, however, that an unencrypted server full of content on-premise is unlikely to be more secure than having the same data encrypted in a best practise cloud environment. Most security specialists agree that building an impenetrable system is impossible, so the pragmatic approach to securing content in the cloud is to have a variety of strategies that can mitigate against an inevitable breach. This can lead to interesting conversations with the software and platform suppliers. For example – assuming that your content is encrypted on the servers of your SaaS provider then who holds the keys to the decryption? Is it the SaaS provider or is it you? There is no correct universal answer to the question, but know that it should be asked and knowing the answer can help to build a risk model and from that a mitigation strategy if you need content security. One of the final issues to consider is “How do you detect a data breach”. This is more difficult than it sounds. If you’ve followed the Linked In 2012 data breach story (see Wikipedia for up to date references) you will see that it took some time to detect that a breach had occurred and much longer to evaluate the full scale of the breach. If security of content is important in your business then detection of a breach must form part of an overall risk mitigation strategy. If you’re new to this topic, then you’ll find all sort of resources available online from blogs like and monthly magazine like all the way through to the Department of Defence Cloud Computing Security Guideline document which provides 152 pages of soporific yet important bedtime reading cloud_computing_srg_v1r1_final.pdf. Time for me to go home and write next month’s class. Now where did I leave my keys?