Benchmarks - Revision history

Daniel-schurhoff-de23@rwth-aachen.de at 13:40, 3 September 2019

2019-09-03T13:40:36Z

Jennifer-witham-0957@rwth-aachen.de at 11:48, 31 January 2019

2019-01-31T11:48:52Z

Nina-loseke-fd7a@rwth-aachen.de at 08:40, 11 January 2019

2019-01-11T08:40:54Z

Nina-loseke-fd7a@rwth-aachen.de at 13:28, 10 January 2019

2019-01-10T13:28:49Z

Jennifer-witham-0957@rwth-aachen.de: Created page with "== General == Benchmarks are applications designed to characterize different performance aspects of a system. The results can then be used to compare and rank different syste..."

2019-01-10T13:11:04Z

Created page with "== General == Benchmarks are applications designed to characterize different performance aspects of a system. The results can then be used to compare and rank different syste..."

New page

== General ==

Benchmarks are applications designed to characterize different performance aspects of a system. The results can then be used to compare and rank different systems (e.g. [https://www.top500.org/ TOP500 list]). Furthermore, they can be used to identify problems and monitor the progress of fixing them. They are also used by vendors to check if the performance requested by the customer is actually achieved. An important aspect for benchmarks is portability, as the main reason for not using "real" appliactions is usually that it takes too long and too many changes to get them to run on a different system. System-related adjustments should be avoided or at least kept to a minimum.

== Benchmark types ==

Individual benchmarks can usually be placed in on one of the following categories:

{| class="wikitable" style="width: 45%;"
| '''Category''' || '''Example''' || '''Description'''
|-
| Raw machine performance || Tflops || machine specifications, e.g. FP peak performance
|-
| Microbenchmarks || Stream || "small" basic components, e.g. latency, bandwidth, communication overhead...
|-
| Algorithmic benchmarks || Linpack || simple algorithms with increasing problem sizes, close to the "kernels"
|-
| Compact Apps/Kernels || MINI-Apps/NPB || more realistic applications that consider more aspects
|-
| Application Suites || SPEC || set of real key applications with different limiting factors
|-
| User-specific applications || Custom benchmarks || run your own application and use hardware counters to manually compute metrics of your choice
|}

← Older revision		Revision as of 13:40, 3 September 2019
Line 1:		Line 1:
		+	[[Category:HPC-User]]
	== General ==		== General ==

← Older revision		Revision as of 08:40, 11 January 2019
Line 22:		Line 22:
	\| User-specific applications \|\| Custom benchmarks \|\| run your own application and use hardware counters to manually compute metrics of your choice		\| User-specific applications \|\| Custom benchmarks \|\| run your own application and use hardware counters to manually compute metrics of your choice
	\|}		\|}
		+
		+	== Resources ==
		+
		+	[https://www.spec.org/ SPEC] (Standard Performance Evaluation Cooperation)
		+
		+	[https://www.cs.virginia.edu/stream/ STREAM] (Sustainable Memory Bandwidth in High Performance Computers)

@@ Line 1: / Line 1: @@
 == General ==
-Benchmarks are applications designed to characterize different performance aspects of a system. The results can then be used to compare and rank different systems (e.g. [https://www.top500.org/ TOP500 list]). Furthermore, they can be used to identify problems and monitor the progress of fixing them. They are also used by vendors to check if the performance requested by the customer is actually achieved. An important aspect for benchmarks is portability, as the main reason for not using "real" appliactions is usually that it takes too long and too many changes to get them to run on a different system. System-related adjustments should be avoided or at least kept to a minimum.
+Benchmarks are applications designed to characterize the performance of a system. The results can then be used to compare and rank different systems (e.g. [https://www.top500.org/ TOP500 list]). Furthermore, they can be used to identify problems and monitor the progress of fixing them. They are also used by vendors to check if the performance requested by the customer is actually achieved. An important aspect for benchmarks is portability, as the main reason for not using "real" appliactions is usually that it takes too long and too many changes to get them to run on a different system. System-related adjustments should be avoided or at least kept to a minimum.
 == Benchmark types ==
@@ Line 7: / Line 9: @@
 Individual benchmarks can usually be placed in one of the following categories:
-{| class="wikitable" style="width: 45%;"
+{| class="wikitable" style="width: 60%;"
 | '''Category''' || '''Example''' || '''Description'''
 |-
 | Microbenchmarks || Stream || "small" basic components, e.g. latency, bandwidth, communication overhead...
 |-
 | Algorithmic benchmarks || Linpack || simple algorithms with increasing problem sizes, close to the "kernels"
 |-
 | Compact Apps/Kernels || MINI-Apps/NPB || more realistic applications that consider more aspects
 |-
-| Application Suites || SPEC || set of real key applications with different limiting factors
+| User-specific applications || Custom benchmarks || run your own application and use hardware counters to manually compute metrics of your choice
 |-
-| User-specific applications || Custom benchmarks || run your own application and use hardware counters to manually compute metrics of your choice
+| Raw machine performance || Tflops || machine specifications, e.g. FP peak performance
 |}
@@ Line 28: / Line 30: @@
 [https://www.cs.virginia.edu/stream/ STREAM] (Sustainable Memory Bandwidth in High Performance Computers)