<b>Cybershuttle: Enabling Scaling of Computational Science Experiments</b>
<p dir="ltr">Computational research is inherently iterative: scientists test multiple models and datasets, each with different data and compute requirements. Within a course of a computational study, there is often a range of models and datasets used, which can vary from small-scale...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | |
| منشور في: |
2025
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| الملخص: | <p dir="ltr">Computational research is inherently iterative: scientists test multiple models and datasets, each with different data and compute requirements. Within a course of a computational study, there is often a range of models and datasets used, which can vary from small-scale (that can be performed in local resources) to large-scale experiments (that require high-performance computing). Scientists across various fields invest considerable effort as the underlying infrastructure changes with the scale of experiments. Cybershuttle aims to provide a solution to this problem, where once a computational experiment is performed, it could be seamlessly scaled across different infrastructures.</p> |
|---|