BookLeaf: An Unstructured Hydrodynamics Mini-application

David Truby; Steven A. Wright; Robert Kevis; Satheesh Maheswaran; Andy Herdman; Stephen A. Jarvis

BookLeaf: An Unstructured Hydrodynamics Mini-application

David Truby, Steven A. Wright, Robert Kevis, Satheesh Maheswaran, Andy Herdman, Stephen A. Jarvis

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

With the age of Exascale computing causing a diversification away from traditional CPU-based homogeneous clusters, it is becoming increasingly difficult to ensure that computationally complex codes are able to run on these emerging architectures. This is especially important for large physics simulations that are themselves becoming increasingly complex and computationally expensive. One proposed solution to the problem of ensuring these applications can run on the desired architectures is to develop representative mini-applications that are simpler and so can be ported to new frameworks more easily, but which are also representative of the algorithmic and performance characteristics of the original applications.
In this paper we present BookLeaf, an unstructured Arbitrary Lagrangian-Eulerian mini-application to add to the suite of representative applications developed and maintained by the UK Mini-App Consortium (UK-MAC). First, we outline the reference implementation of our application in Fortran. We then discuss a number of alternative implementations using a variety of parallel programming models and discuss the issues that arise when porting such an application to new architectures. To demonstrate our implementation, we present a study of the performance of BookLeaf on number of platforms using alternative designs, and we document a scaling study showing the behaviour of the application at scale.

Original language	English
Number of pages	8
Publication status	Published - 10 Sept 2018
Event	IEEE International Conference on Cluster Computing 2018 - Belfast, United Kingdom Duration: 10 Sept 2018 → 13 Oct 2018 Conference number: 20th https://cluster2018.github.io/

Conference

Conference	IEEE International Conference on Cluster Computing 2018
Abbreviated title	CLUSTER
Country/Territory	United Kingdom
City	Belfast
Period	10/09/18 → 13/10/18
Internet address	https://cluster2018.github.io/

Access to Document

BookLeaf-unstructured-hydrodynamics-mini-application-Truby-2018Accepted author manuscript, 271 KB

Cite this

@conference{ec63dc95720843fbbe8469590aa45d9b,

title = "BookLeaf: An Unstructured Hydrodynamics Mini-application",

abstract = "With the age of Exascale computing causing a diversification away from traditional CPU-based homogeneous clusters, it is becoming increasingly difficult to ensure that computationally complex codes are able to run on these emerging architectures. This is especially important for large physics simulations that are themselves becoming increasingly complex and computationally expensive. One proposed solution to the problem of ensuring these applications can run on the desired architectures is to develop representative mini-applications that are simpler and so can be ported to new frameworks more easily, but which are also representative of the algorithmic and performance characteristics of the original applications.In this paper we present BookLeaf, an unstructured Arbitrary Lagrangian-Eulerian mini-application to add to the suite of representative applications developed and maintained by the UK Mini-App Consortium (UK-MAC). First, we outline the reference implementation of our application in Fortran. We then discuss a number of alternative implementations using a variety of parallel programming models and discuss the issues that arise when porting such an application to new architectures. To demonstrate our implementation, we present a study of the performance of BookLeaf on number of platforms using alternative designs, and we document a scaling study showing the behaviour of the application at scale.",

author = "David Truby and Wright, {Steven A.} and Robert Kevis and Satheesh Maheswaran and Andy Herdman and Jarvis, {Stephen A.}",

year = "2018",

month = sep,

day = "10",

language = "English",

note = "IEEE International Conference on Cluster Computing 2018, CLUSTER ; Conference date: 10-09-2018 Through 13-10-2018",

url = "https://cluster2018.github.io/",

}

TY - CONF

T1 - BookLeaf: An Unstructured Hydrodynamics Mini-application

AU - Truby, David

AU - Wright, Steven A.

AU - Kevis, Robert

AU - Maheswaran, Satheesh

AU - Herdman, Andy

AU - Jarvis, Stephen A.

N1 - Conference code: 20th

PY - 2018/9/10

Y1 - 2018/9/10

N2 - With the age of Exascale computing causing a diversification away from traditional CPU-based homogeneous clusters, it is becoming increasingly difficult to ensure that computationally complex codes are able to run on these emerging architectures. This is especially important for large physics simulations that are themselves becoming increasingly complex and computationally expensive. One proposed solution to the problem of ensuring these applications can run on the desired architectures is to develop representative mini-applications that are simpler and so can be ported to new frameworks more easily, but which are also representative of the algorithmic and performance characteristics of the original applications.In this paper we present BookLeaf, an unstructured Arbitrary Lagrangian-Eulerian mini-application to add to the suite of representative applications developed and maintained by the UK Mini-App Consortium (UK-MAC). First, we outline the reference implementation of our application in Fortran. We then discuss a number of alternative implementations using a variety of parallel programming models and discuss the issues that arise when porting such an application to new architectures. To demonstrate our implementation, we present a study of the performance of BookLeaf on number of platforms using alternative designs, and we document a scaling study showing the behaviour of the application at scale.

AB - With the age of Exascale computing causing a diversification away from traditional CPU-based homogeneous clusters, it is becoming increasingly difficult to ensure that computationally complex codes are able to run on these emerging architectures. This is especially important for large physics simulations that are themselves becoming increasingly complex and computationally expensive. One proposed solution to the problem of ensuring these applications can run on the desired architectures is to develop representative mini-applications that are simpler and so can be ported to new frameworks more easily, but which are also representative of the algorithmic and performance characteristics of the original applications.In this paper we present BookLeaf, an unstructured Arbitrary Lagrangian-Eulerian mini-application to add to the suite of representative applications developed and maintained by the UK Mini-App Consortium (UK-MAC). First, we outline the reference implementation of our application in Fortran. We then discuss a number of alternative implementations using a variety of parallel programming models and discuss the issues that arise when porting such an application to new architectures. To demonstrate our implementation, we present a study of the performance of BookLeaf on number of platforms using alternative designs, and we document a scaling study showing the behaviour of the application at scale.

M3 - Paper

T2 - IEEE International Conference on Cluster Computing 2018

Y2 - 10 September 2018 through 13 October 2018

ER -