Performance of the System

9.0Performance of the System

Sequoia benchmarks serve three purposes in the Sequoia procurement. First, the benchmarks provide the Offeror with the opportunity to provide LLNS concrete data (in its RFP response) associated with the performance, reliability and scalability of the proposed systems on programmatically important applications. Offeror’s proposal response should include information of many kinds, including a report on the results of running Sequoia benchmarks on existing Offeror hardware, simulators and extrapolations to future proposed systems. In this role, the benchmarks play an essential role in the proposal evaluation process. Second, the benchmarks will be used as an integral part of the systems SWL tests (see Section 8.3.3) in order to assess the proposed systems ability to meet or exceed Sequoia performance, scalability and stability technical requirements. Third, these benchmarks will be used to assess continuous performance improvement of compiler and other critical software technologies during development and after acceptance over the lifetime of the Dawn and Sequoia systems.

While performance is an important consideration in any computer acquisition, for the ASC Program it matters greatly how improved performance is achieved. See Section 9.4.1 for a discussion of allowed modifications to the Sequoia benchmarks.

The Sequoia benchmarks described below may be executed by the Offeror and results presented in its proposal response for the purpose of measuring the execution performance and compiler capabilities of existing systems that may be proposed. In addition, these benchmark results from existing systems may be used by Offeror to extrapolate and/or estimate the benchmark performance on future proposed systems. The general requirements and constraints outlined below apply to all of the benchmark codes. Additional requirements and/or constraints found in individual benchmark readme files apply to that individual benchmark.