Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

adapt Oceanstor pacific 100D filesystem #23

Open
wants to merge 1 commit into
base: huawei
Choose a base branch
from

Conversation

openmpi
Copy link

@openmpi openmpi commented Jun 11, 2021

hypermpi adapt OceanStor Pacific 100D filesystem,

@CLAassistant
Copy link

CLAassistant commented Jun 11, 2021

CLA assistant check
All committers have signed the CLA.

@openmpi openmpi changed the base branch from master to v4.0.x June 16, 2021 02:30
@openmpi openmpi changed the base branch from v4.0.x to v4.1.x June 16, 2021 02:34
@openmpi openmpi changed the base branch from v4.1.x to huawei June 16, 2021 02:54
@openmpi openmpi force-pushed the Oceanstor_Pacific_100D_MPI branch 6 times, most recently from 862df3d to eda01ea Compare June 22, 2021 03:40
@openmpi openmpi force-pushed the Oceanstor_Pacific_100D_MPI branch 4 times, most recently from 43f3755 to 9d6e1bb Compare July 7, 2021 08:31
* zero, it's probably still a good idea to mix the "small" file domains
* across the fd_size array to keep the io nodes in balance */
ADIO_Offset n_oceanfs_blk = fd_oceanfs_range / blksize;
ADIO_Offset nb_cn_small = n_oceanfs_blk / naggs;

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If possible, I think it would make sense to reduce the use of division and multiplication operations and re-use results when possible, such as (n_oceanfs_blk / naggs).
This would yield better performance on ARM platforms as well as Intel.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you for your suggestions. I also checked other codes to minimize the calculation of multiplication and division;


for (i = 0; i < naggs; i++) {
if (i < naggs_large) {
fd_size[i] = (nb_cn_small + 1) * blksize;

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since (nb_cn_small + 1) * blksize; and (nb_cn_small * blksize) are constant values in the loop, they can be pre-calculated and assigned into 2 variables that will be used inside the loop.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(nb_cn_small + 1) * blksize; and (nb_cn_small * blksize) are constant values,I have defined variables to save the results to avoid repeated calculations in the loop;

@openmpi openmpi force-pushed the Oceanstor_Pacific_100D_MPI branch from 3acfd56 to 90e5257 Compare August 21, 2021 10:11
@it-is-a-robot
Copy link

Thanks for your pull request. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

The following commits have not yet signed CLA.

90e5257 | adapt OceanStor Pacipic 100D filesystem

📝 Please access here to sign the CLA.

It may take a couple minutes for the CLA signature to be fully registered; after that, please reply here with a new comment: /check-cla to verify. Thanks.


  • If you've already signed a CLA, it's possible you're using a different email address for your gitee account. Check your existing CLA data and verify the email.
  • If you signed the CLA as an employee or a member of an organization, please contact your corporation or organization to verify you have been activated to start contributing.
  • If you have done the above and are still having issues with the CLA being reported as unsigned, please feel free to file an issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants