Performance: Why do large equilibrium splines slow down the code?

未关闭
#376 4 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 3 天内回复

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
42/100
Issue 类型
缺陷
描述清晰度
基本清楚
活跃度
活跃
技术栈
julia
领域
performance

调研方向

Start by reading the spline evaluation calls made during ForceFreeStates ODE integration, especially how knot hints are used as mpsi grows. Compare performance across equilibrium sizes to identify what causes the slowdown. Done means explaining the scaling and showing whether an optimization makes runtime less sensitive to knot count.

由索引模型根据 Issue 内容生成。

描述

perf

Empirically, we all know larger mpsi slows down the ForceFreeStates.

The hand waving reason is the 10s of thousands of spline interpolations done during the ODE integration are a bit slower. But why would this be the case when we are using hints to cleanly provide the knot locations of interest and these are only slowly (and monotonically) incrementing? It seems like there would be effectively no search difference with knot number in this case. Is there some further optimization of the spline evaluation calls that would make this loop insensitive to the spline knot length? It seems like there should be.

@mgyoo86 might have some insights.

Possible topic for David Smitth ( @drsmith48 ) and UW hackathon

主要语言
Julia
星标
11
派生
7
平均合并
4 天 3 小时
30 天内合并 PR
70

环境准备

  • 没有 Dockerfile 或 Docker Compose 文件
  • 有 Pull Request 模板
  • 没有贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

OpenFUSIONToolkit/GPEC 的其他 Issue

查看 OpenFUSIONToolkit/GPEC 的全部 Issue

相似的 Issue

更多 Julia Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。