Presentation Proposal for BlackHat 2021
At the outset, Time Locks and Time Series Databases are two extreme ends. However there are interesting possibilities to bring them together. In this session, I am presenting a creative convergence of post quantum cryptography powered time locks engineered to secure and scale multi party computation on time series databases in a privacy preserving manner.
The presentation will consist of sections on time locks and their different manifestations such as verified delay functions, verified random functions, time aggregators and time clock types. The next section will be on the types of time series databases. This section will also cover the core functional and non-functional aspects of time series databases. The subsequent section will outline how different types of recursions are implemented on time lock functions. Then there will be a reference architecture of how time locks can secure time series databases.
This concept of time locks and time series databases is an evolution of my research on time lock machines and manifestations ever since working on quantum and classical randomness. I had presented a session on the application of time locks in DEFCON 28 to construct verified randomness for distributed ledgers such as Ethereum, Algorand, DFNITY etc. I have worked further on time locks and post quantum cryptography ever since the DEFCON 28. In the NullCon 28, I had presented a session on the emerging ensembles of post quantum cryptography.
The current proposal is an innovation invocation of time locks in the time series databases with first of its kind algorithm for sharding and scaling time series databases with time locks in a recursive and recurrent manner. This session is designed to deliver deep insights into the security and scalability aspects of time series databases and the potentials and promises of time locks. This session will also help the audience to understand how the creative convergence of time series databases and time locks can produce results in multi party computation setting.
My research is an attempt to address the issues of clutter and chaos in time series analysis which becomes a major impediment for privacy preserving federated machine learning experiments and secure and scalable multi party computations essential for analytic use cases in GovTech, HealthTech, EnergyTech, FinTech and RegTech industrial sectors. At the moment, there are very limited implementations of privacy preserving tools on time series databases. Thus my concept helps to address the security and scalability challenges of time series databases in a multi party computation setting.
I have identified a number of vulnerabilities in time series databases used in energy and utility domain, fintech domain and healthtech domain. These vulnerabilities are due to the lack of time locks in the distributed systems and satellites used to generate the metrics of time in the synchronous, and asynchronous computational processes.