Paper Title: Validating and improving QLC enterprise controllers with data-driven development

Paper Abstract: During recent years, 3D QLC NAND flash has been successfully established in the enterprise storage market. A wide range of innovative mitigation techniques have been proposed to address the peculiarities of this non-volatile memory generation. But which of these techniques are essential to improve overall performance in terms of bandwidth and latency? And how can the performance be sustained in presence of short term bursts as well as over longer time periods? We review various data placement strategies in a hybrid controller architecture that uses a variable amount of the QLC blocks in SLC mode and present evidence from their effectiveness using field data. While modeling, simulation, and wide-range testing are essential to validate the efficiency of data placement strategies, they cannot predict all potential interactions of workloads that may be seen in reality. We demonstrate that data-driven development can be a key contributor to address this increasing complexity.

Paper Author: Roman Pletka, Research Staff Member, IBM Research

Author Bio: Roman Pletka is a research staff member and master inventor for cloud storage, data, and AI systems at the IBM Zurich Research Laboratory where he focuses on non-volatile memory technologies in storage systems. He is a frequent speaker at international conferences, has published over 20 articles and obtained more than 100 patents in managing non-volatile memories, security, scalability, and availability of distributed storage systems as well as quality-of-service in high-speed networks, active networks, and network processors. He has made presentations at many international conferences including the ACM International Conference on Systems and Storage (SYSTOR) and the Nonvolatile Memory Workshop. He has over 17 years experience in storage systems research. He earned a PhD in computer networking from ETH Zurich, Switzerland and an MS in the same subject from EPFL (Swiss Federal Institute of Technology of Lausanne).