Big data technology and implications for security research

‘Big data’ has become a buzzword that is increasingly vague. Attempts to nail it down to the attributes of volume, velocity, and variety are helpful but have shortcomings. In particular, these widely used terms are attributes and not definitions in and of themselves. The question is what assignment or combination of these attributes creates big data. What are the theoretical limitations on big data computation? From this analysis, can we identify taxonomies of big data problems? This talk seeks to shed light on these questions by analyzing the computer science problems and limitations which initially formed the big data movement. It then uses those insights to evaluate how to best apply big data approaches to computer security. We discuss how big data might open new avenues for security research, how big data is enhancing commercial security tools, and the state of security in existing big data products.