TU Darmstadt / ULB / TUbiblio

Learning Cost-Effective Sampling Strategies for Empirical Performance Modeling

Ritter, Marcus ; Calotoiu, Alexandru ; Rinke, Sebastian ; Reimann, Thorsten ; Hoefler, Torsten ; Wolf, Felix (2020)
Learning Cost-Effective Sampling Strategies for Empirical Performance Modeling.
34th IEEE International Parallel and Distributed Processing Symposium (IPDPS'20). New Orleans, USA (18.-22.05.2020)
doi: 10.1109/IPDPS47924.2020.00095
Conference or Workshop Item, Bibliographie

Abstract

Identifying scalability bottlenecks in parallel applications is a vital but also laborious and expensive task. Empirical performance models have proven to be helpful to find such limitations, though they require a set of experiments in order to gain valuable insights. Therefore, the experiment design determines the quality and cost of the models. Extra-P is an empirical modeling tool that uses small-scale experiments to assess the scalability of applications. Its current version requires an exponential number of experiments per model parameter. This makes the creation of empirical performance models very expensive, and in some situations even impractical. In this paper, we propose a novel parameter-value selection heuristic, which functions as a guideline for the experiment design, leveraging sparse performance-modeling, a technique that only needs a polynomial number of experiments per model parameter. Using synthetic analysis and data from three different case studies, we show that our solution reduces the average modeling costs by about 85% while retaining 92% of the model accuracy.

Item Type: Conference or Workshop Item
Erschienen: 2020
Creators: Ritter, Marcus ; Calotoiu, Alexandru ; Rinke, Sebastian ; Reimann, Thorsten ; Hoefler, Torsten ; Wolf, Felix
Type of entry: Bibliographie
Title: Learning Cost-Effective Sampling Strategies for Empirical Performance Modeling
Language: English
Date: 14 July 2020
Publisher: IEEE
Book Title: Proceedings: 2020 IEEE 34th International Parallel and Distributed Processing Symposium
Event Title: 34th IEEE International Parallel and Distributed Processing Symposium (IPDPS'20)
Event Location: New Orleans, USA
Event Dates: 18.-22.05.2020
DOI: 10.1109/IPDPS47924.2020.00095
Abstract:

Identifying scalability bottlenecks in parallel applications is a vital but also laborious and expensive task. Empirical performance models have proven to be helpful to find such limitations, though they require a set of experiments in order to gain valuable insights. Therefore, the experiment design determines the quality and cost of the models. Extra-P is an empirical modeling tool that uses small-scale experiments to assess the scalability of applications. Its current version requires an exponential number of experiments per model parameter. This makes the creation of empirical performance models very expensive, and in some situations even impractical. In this paper, we propose a novel parameter-value selection heuristic, which functions as a guideline for the experiment design, leveraging sparse performance-modeling, a technique that only needs a polynomial number of experiments per model parameter. Using synthetic analysis and data from three different case studies, we show that our solution reduces the average modeling costs by about 85% while retaining 92% of the model accuracy.

Uncontrolled Keywords: LOEWE|SF4.0, DFG|323299120, DFG|320898076, BMBF|01IH16008D, DoE|DE-SC0015524, DFG, BMBF, LOEWE
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Parallel Programming
Zentrale Einrichtungen
Zentrale Einrichtungen > University IT-Service and Computing Centre (HRZ)
Zentrale Einrichtungen > University IT-Service and Computing Centre (HRZ) > Hochleistungsrechner
Date Deposited: 04 Apr 2024 09:04
Last Modified: 04 Apr 2024 09:04
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details