9 Aug 2023
#CodeePerformanceTip
Performance can be increased by using the highest possible trip count in the vectorized loop.
Consider interchanging loops so that the loop with the highest trip count becomes the innermost loop.
Learn more: codee.com/catalog/pwr019/
2
3
111
1 Aug 2023
#CodeePerformanceTip Avoid unnecessary floating-point data conversions involving constants.
Instead of the double constant, explicitly specify a float constant by appending the constant with f.
Learn more: codee.com/catalog/pwr044/
1
2
85
1 Jun 2023
#CodeePerformanceTip
Each procedure parameter should have its intent declared to facilitate reasoning about the flow of data in and out of the function.
Read more: codee.com/knowledge/pwr008/
#Fortran #C #BestPractices #Performance
2
2
147
25 May 2023
#CodeePerformanceTip
Explicitly declare all the global variables read or written in the function either as parameters or as local variables in the function body, including the output variables.
🔗Read more: codee.com/knowledge/pwr001/
#Fortran #C #BestPractices #Performance
2
2
125
16 May 2023
#CodeePerformanceTip: Consider loop fission to enable #vectorization.
By extracting the vectorizable computations to a new loop, it can become vectorized with the corresponding #performance increase.
👉codee.com/knowledge/pwr020/
#BestPractices #Codee #automation
3
2
154
3 May 2023
#CodeePerformanceTip Consider applying #multithreading parallelism to scalar reduction loop.
You must specify how to execute the loop in vector mode on the hardware, as well as add the appropriate synchronization to avoid race conditions at runtime.
🔗 codee.com/knowledge/pwr051/
2
2
200
19 Apr 2023
#CodeePerformanceTip
Consider loop interchange to improve the locality of reference and enable vectorization.
Using loop interchange, the inefficient matrix access pattern is replaced with a more efficient one.
See the explanation: 🔗zurl.co/uyjR
#performance
4
4
305
5 Apr 2023
#CodeePerformanceTip
Explicitly declaring scalar variables in the smallest scope possible makes it easier for the compiler to track its usage.
🔗Read more: codee.com/knowledge/pwr002/
#performance #software #bestpractices
6
8
373
28 Mar 2023
#CodeePerformanceTip
The most efficient way to process #arrays is to iterate over its elements in the same order in which they are laid out in memory, so that the program performs a sequential access to consecutive data in memory.
👉 codee.com/knowledge/pwr010/
#C #Cpp #performance
1
2
3
242
27 Sep 2022
#CodeePerformanceTip: Your loop is rereading the same data from the memory many times? Have you tried loop tiling?
👇 From our knowledge base about when and how to apply loop tiling: codee.com/knowledge/glossary…
#looptiling #performance #software #HPC #Codee"
1
2
11 Aug 2022
#CodeePerformanceTip: Does your computation need to be both precise and fast?
Read more about how to improve the performance of your computations without affecting the speed in the post: codee.com/run-floating-point…
1
3
14 Jun 2022
#CodeePerformanceTip: In many applications, the CPU has to wait for the data from the memory. But is this our destiny? Can we do something to improve on this?
From our blog about #MemoryEfficiency and other ways to improve the performance of your code: codee.com/many-ways-to-speed… 👇
1
2
7 Jun 2022
#CodeePerformanceTip: When optimizing nested loops and matrix code, what are the best practices?
👇From our blog post about loop interchange and guidelines for good performance: codee.com/case-study-how-we-…
1
5
24 May 2022
#CodeePerformanceTip: Does your computation need to be both precise and fast?
👇 From our blog about how to improve the performance of your computations without affecting the speed: codee.com/run-floating-point…
1
2
10 May 2022
#CodeePerformanceTip: How to determine if your important algorithm is running at peak performance, i.e. it is using the hardware resources efficiently?
From our blog about the roofline model: codee.com/is-your-algorithm-… 👇
3
5
5 May 2022
#CodeePerformanceTip: Performance improvements are possible by distributing the workload to accelerators, such as GPUs.
From our blog about many ways to speed up your program in the post: codee.com/many-ways-to-speed…👇
2
2
3 May 2022
#CodeePerformanceTip: Performance and numerical precision, do they go together?
From our blog about portable compiler pragmas and how you can use them to improve the speed of your program with minimal reduction in performance: codee.com/trade-precision-fo… 👇
1
2
28 Apr 2022
#CodeePerformanceTip: In many applications, the CPU has to wait for the data from the memory. But is this our destiny? Can we do something to improve on this?
From our blog about memory efficiency and other ways to improve the performance of your code: codee.com/many-ways-to-speed… 👇
3
3
26 Apr 2022
#CodeePerformanceTip: When optimizing nested loops and matrix code, what are the best practices?
From our blog post about loop interchange and guidelines for good performance 🔗 codee.com/case-study-how-we-… 👇
1
2