Keita Iwabuchi
Lawrence Livermore National Laboratory
We present Metall, a persistent memory allocator designed to provide developers with an API to allocate custom C++ data structures in both block-storage and byte-addressable persistent memories (e.g., NVMe and Intel Optane DC Persistent Memory). Metall incorporates the state-of-the-art allocation algorithms in Supermalloc with the rich C++ interface developed by Boost.Interprocess, and provides persistent memory snapshotting (versioning) capabilities. We perform a performance evaluation using a graph construction benchmark, persistently storing data structures on NVMe SSDs and byte-addressable persistent memory. Our approach using Metall works without code modifications in both sceneries — block-addressable and byte-addressable. Metall provides a coarse-grained consistency model, allowing the application to determine when it is appropriate to create durable snapshots of the persistent heap.
Dr. Keita Iwabuchi is a data scientist in the Center for Applied Scientific Computing (CASC) at Lawrence Livermore National Laboratory. His research area is distributed systems and parallel computing, particularly in high performance computing (HPC). Major focuses are HPC-scale graph analytics and system software for persistent memory and non-volatile memory.
Dr. Iwabuchi received his Ph.D. in Mathematical and Computing Sciences from Tokyo Institute of Technology in 2017. He joined CASC as a postdoctoral researcher in 2017, and became a staff scientist in 2020.
Dr. Iwabuchi received his Ph.D. in Mathematical and Computing Sciences from Tokyo Institute of Technology in 2017. He joined CASC as a postdoctoral researcher in 2017, and became a staff scientist in 2020.
Where/When
Time: Oct 19, 2020 8:00 PM – 9:00 PM Pacific Time (US and Canada)
Zoom link: https://ucsd.zoom.us/j/91478040499