Energy, medical, aerospace, weather, entertainment (to name just a few), have all, over the years, acquired and recorded massive volumes of raw data. The challenge they have continued to face is to determine what data is useful (solving a real problem, fulfilling a real need, or monetizable), and what data has little to no value and must be discarded. If re-acquiring the data is practically or financially prohibitive, it probably means multiple copies are retained in the event of the original being destroyed.

The panel will review these questions with an aim to list some general considerations that may assist Data Managers in their operational discussions with their data owners.

With the O&G industry expanding into renewable energy technologies, what are the ramifications for computing and data infrastructure? E&P has traditionally required large HPC (High Performance Computing) clusters to run large simulations.  Renewable energy workloads often rely more on AI/ML techniques and different types of simulation.

The good news is the architecture required for HPC implementations has many similarities with AI implementations. Both use high levels of compute and storage, large memory capacity and bandwidth, and high-bandwidth fabrics to achieve results, typically by processing massive data sets of increasing size. Both have been leveraging accelerated computing approaches, typically using GPU technology.