Re: [PATCH 0/2] nvmet: support polling task for RDMA and TCP

[Ping Gan <jacky_gam_2001@163.com>]

> On 7/4/24 10:10, Ping Gan wrote:
>>> On 02/07/2024 13:02, Ping Gan wrote:
>
>>>> And the bandwidth of a node is only 3100MB. While we used the patch
>>>> and enable 6 polling task, the bandwidth can be 4000MB. It's a good
>>>> improvement.
>>>
>>> I think you will see similar performance with unbound workqueue and
>>> rps.
>> 
>> Yes, I remodified the nvmet-tcp/nvmet-rdma code for supporting
>> unbound
>> workqueue, and in same prerequisites of above to run test, and
>> compared
>> the result of unbound workqueue and polling mode task. And I got a
>> good
>> performance for unbound workqueue. For unbound workqueue TCP we got
>> 3850M/node, it's almost equal to polling task. And also tested
>> nvmet-rdma we get 5100M/node for unbound workqueue RDMA versus 5600M
>> for
>> polling task, seems the diff is very small. Anyway, your advice is
>> good.
>> Do you think we should submit the unbound workqueue patches for
>> nvmet-tcp
>> and nvmet-rdma to upstream nvmet?
>
> Please do. I have been using pretty much the same patch during
> development of my nvme-tcp scalability patchset, and using WQ_UNBOUND
> definitely improves the situation here.

Thanks for your confirm! Okay, will do that.

Regards,
Ping