Specifications include, but are not limited to: i)All proposed hardware, software (released commercially or as open source), and infrastructure must be supported by the Supplier at the time of system acceptance and throughout the subsequent 48 months of production system use. ii)The proposed system must be a balanced, commercially-available, production-grade HPC system that contains an appropriate combination of processor, memory, interconnect, disk input/output (I/O), and operating system (OS) capabilities in order to execute complex, tightly-coupled, large-scale, scientific calculations; more specifically, the system must be able to successfully execute a variety of workloads, including jobs which stress all subsystems and which require the simultaneous, tightly-coupled use of the full number of compute nodes within the system. iii)All equipment must be new; no refurbished or used equipment is allowed. iv)All nodes and operating systems that are provided must include any required fixes for Spectre/Meltdown vulnerabilities. v)The entire system, including all compute nodes, support nodes (login, data transfer, system management), networking, and storage must fit into no more than 14 racks (1)The maximum linear floor space available for contiguous racks (maximum row-length) is 20 feet. (2)All racks containing compute nodes shall be configured in a contiguous manner. (3)The storage systems racks will not be located adjacent to the compute node racks, but instead will be elsewhere within the data center. The storage system racks will be within 100 feet of the compute node racks; the exact location of these racks will be determined post-award.