objectwriter: Allow customize maxChunkLen #96

Merged
fyrchik merged 1 commit from dstepanov-yadro/frostfs-sdk-go:feat/customize_max_chunk_len into master 2023-06-27 07:01:11 +00:00

Closes #80

  1. We can create some handler on frostfs-node to get server parameters. But actually we shoul ask every node from netmap about max send/receive size.

  2. Now there is no option to change message size on node, so looks like this is not required now.

Closes #80 1. We can create some handler on frostfs-node to get server parameters. But actually we shoul ask every node from netmap about max send/receive size. 2. Now there is no option to change message size on node, so looks like this is not required now.
dstepanov-yadro requested review from storage-core-committers 2023-06-23 14:33:04 +00:00
dstepanov-yadro requested review from storage-core-developers 2023-06-23 14:33:05 +00:00
fyrchik changed title from objectwriter: Allow custimize maxChunkLen to objectwriter: Allow customize maxChunkLen 2023-06-25 10:45:12 +00:00
acid-ant approved these changes 2023-06-26 05:55:56 +00:00
dstepanov-yadro force-pushed feat/customize_max_chunk_len from 4949931c82 to fed623ed91 2023-06-26 06:50:28 +00:00 Compare
fyrchik reviewed 2023-06-26 09:29:14 +00:00
@ -42,0 +44,4 @@
// Maximum chunk length restricts maximum byte length of the chunk
// transmitted in a single stream message. It depends on
// server settings and other message fields.
// If not specified or negative value set, default value of 3MB will be used.
Owner

s/MB/MiB/ ?

s/MB/MiB/ ?
Author
Member

fixed

fixed
fyrchik marked this conversation as resolved
@ -42,0 +45,4 @@
// transmitted in a single stream message. It depends on
// server settings and other message fields.
// If not specified or negative value set, default value of 3MB will be used.
func (x *PrmObjectPutInit) SetMaxChunkLen(v int) {
Owner

I am now thinking: this setting is somewhat low-level and relates only to the gRPC. It is also confusing because we have MaxObjectSize in #94/ , which is also MaxChunkLen in some sense.

May we somehow reflect this in naming? cc @acid-ant

My suggestion would be:

  • SetgRPCPayloadChunkLen (longer name communicates the complexity and creates cognitive load)
  • SetObjectPartSize for #94

We surely can come up with something better. @TrueCloudLab/storage-core-committers @TrueCloudLab/storage-core-developers

I am now thinking: this setting is somewhat low-level and relates only to the gRPC. It is also confusing because we have `MaxObjectSize` in https://git.frostfs.info/TrueCloudLab/frostfs-sdk-go/pulls/94/ , which is also `MaxChunkLen` in some sense. May we somehow reflect this in naming? cc @acid-ant My suggestion would be: - `SetgRPCPayloadChunkLen` (longer name communicates the complexity and creates cognitive load) - `SetObjectPartSize` for #94 We surely can come up with something better. @TrueCloudLab/storage-core-committers @TrueCloudLab/storage-core-developers
Author
Member

fixed

fixed
Member

I agree, that it is better to reflect in naming that this parameter relies on gRPC.

I agree, that it is better to reflect in naming that this parameter relies on gRPC.
fyrchik marked this conversation as resolved
dstepanov-yadro force-pushed feat/customize_max_chunk_len from fed623ed91 to c243b443bc 2023-06-26 15:13:22 +00:00 Compare
fyrchik approved these changes 2023-06-27 07:00:59 +00:00
fyrchik merged commit c243b443bc into master 2023-06-27 07:01:11 +00:00
Sign in to join this conversation.
No reviewers
TrueCloudLab/storage-core-developers
No milestone
No project
No assignees
3 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: TrueCloudLab/frostfs-sdk-go#96
No description provided.