【24h】

LSPR Benchmark Converter

机译:LSPR基准转换器

获取原文
获取原文并翻译 | 示例
获取外文期刊封面目录资料

摘要

A pair of trained neural networks are used as mapping functions to translate the published LSPR benchmarks between the zOS 1.4 and OS/390 v2.10 benchmark sets. The z890 and z990 machines are not represented in the OS/390 v2.10 benchmarks, and the neural networks provide an estimate of the z890 and z990 performance under the old benchmarks. Capacity planners can provide a capacity estimate of these processors directly comparable to the older pre-zSeries machines that are being replaced by using previously validated LSPR workload distributions along with these neural networks. The z800 and z900 machines are present in both sets of benchmarks and are used as the training sets for the neural networks.
机译:一对经过训练的神经网络用作映射功能,可以在zOS 1.4和OS / 390 v2.10基准测试集之间转换已发布的LSPR基准测试。 z890和z990机器没有出现在OS / 390 v2.10基准测试中,并且神经网络提供了在旧基准下对z890和z990性能的估计。容量规划人员可以提供这些处理器的容量估计,这些容量估计可与以前的zSeries之前的旧机器相比,后者将通过使用先前验证的LSPR工作负载分布以及这些神经网络进行替换。 z800和z900机器同时存在于这两组基准中,并用作神经网络的训练集。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号