No abstract

No abstract

HPC systems are a critical resource for scientific research and advanced industries. The demand for computational power and memory is increasing and ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of many compute nodes and are consequently expected to experience frequent faults and crashes. Exact state reconstruction (ESR) has been proposed as a mechanism to alleviate the impact of frequent failures on long-term computations. ESR has shown great potential in the context of iterative linear algebra solvers, a key building block in numerous scientific applications.Recent designs of supercomputers feature the emerging nonvolatile memory (NVM) technology. For example, the Exascale Aurora supercomputer is planned to integrate Intel Optane™ DCPMM. This work investigates how NVM can be used to improve ESR so that it can scale to future exascale systems such as Aurora and provide enhanced resilience.We propose the non-volatile memory ESR (NVM-ESR) mechanism. NVM-ESR demonstrates how NVM can be utilized in supercomputers for enabling efficient recovery from faults while requiring significantly smaller memory footprint and time overheads in comparison to ESR. We focus on the preconditioned conjugate gradient (PCG) iterative solver also studied in prior ESR research, because it is employed by the representative HPCG scientific benchmark.The source code used by this work, as well as the benchmarks and other relevant sources, are available at: https://github.com/ Scientific-Computing-Lab-NRCN/NVM-ESR.git.

As the High Performance Computing (HPC) world moves towards the Exa-Scale era, huge amounts of data should be analyzed, manipulated and stored. In the traditional storage/memory hierarchy, each compute node retains its data objects in its local volatile DRAM. Whenever the DRAM's capacity becomes insufficient for storing this data, the computation should either be distributed between several compute nodes, or some portion of these data objects must be stored in a non-volatile block device such as a hard disk drive (HDD) or an SSD storage device. These standard block devices offer large and relatively cheap non-volatile storage, but their access times are orders-ofmagnitude slower than those of DRAM. Optane™ DataCenter Persistent Memory Module (DCPMM) [1], a new technology introduced by Intel, provides non-volatile memory that can be plugged into standard memory bus slots (DDR DIMMs) and therefore be accessed much faster than standard storage devices. In this work, we present and analyze the results of a comprehensive performance assessment of several ways in which DCPMM can 1) replace standard storage devices, and 2) replace or augment DRAM for improving the performance of HPC scientific computations. To achieve this goal, we have configured an HPC system such that DCPMM can service I/O operations of scientific applications, replace standard storage devices and file systems (specifically for diagnostics and checkpoint-restarting), and serve for expanding applications' main memory. We focus on keeping the scientific codes with as few changes as possible, while allowing them to access the NVM transparently as if they access persistent storage. Our results show that DCPMM allows scientific applications to fully utilize nodes' locality by providing them with sufficiently-large main memory. Moreover, it can also be used for providing a high-performance replacement for persistent storage. Thus, the usage of DCPMM has the potential of replacing standard HDD and SSD storage devices in HPC architectures and enabling a more efficient platform for modern supercomputing applications. The source code used by this work, as well as the benchmarks and other relevant sources, are available at: https://github.com/ Scientific-Computing-Lab-NRCN/StoringStorage.