Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-12 Thread Michal Rostecki
Hi Anand, re: inflight calculation On Thu, Feb 11, 2021 at 03:55:33PM +, Michal Rostecki wrote: > > It is better to have random workloads in the above three categories > > of configs. > > > > Apart from the above three configs, there is also > > all-non-rotational with hetero > > For exampl

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-11 Thread Michal Rostecki
On Wed, Feb 10, 2021 at 04:20:20PM +0800, Anand Jain wrote: > On 10/02/2021 04:30, Michal Rostecki wrote: > > The penalty value is an additional value added to the number of inflight > > requests when a scheduled request is non-local (which means it would > > start from the different physical locat

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-11 Thread Michal Rostecki
On Thu, Feb 11, 2021 at 03:27:38AM +0100, Michał Mirosław wrote: > On Wed, Feb 10, 2021 at 07:23:04PM +, Michal Rostecki wrote: > > On Wed, Feb 10, 2021 at 01:58:15PM +0100, Michał Mirosław wrote: > > > On Wed, Feb 10, 2021 at 12:29:25PM +, Michal Rostecki wrote: > > > > On Wed, Feb 10, 202

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-10 Thread Michał Mirosław
On Wed, Feb 10, 2021 at 07:23:04PM +, Michal Rostecki wrote: > On Wed, Feb 10, 2021 at 01:58:15PM +0100, Michał Mirosław wrote: > > On Wed, Feb 10, 2021 at 12:29:25PM +, Michal Rostecki wrote: > > > On Wed, Feb 10, 2021 at 05:24:28AM +0100, Michał Mirosław wrote: > > > > This looks like it

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-10 Thread Michal Rostecki
On Wed, Feb 10, 2021 at 01:58:15PM +0100, Michał Mirosław wrote: > On Wed, Feb 10, 2021 at 12:29:25PM +, Michal Rostecki wrote: > > On Wed, Feb 10, 2021 at 05:24:28AM +0100, Michał Mirosław wrote: > > > This looks like it effectively decreases queue depth for non-last > > > device. After all de

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-10 Thread Michał Mirosław
On Wed, Feb 10, 2021 at 12:29:25PM +, Michal Rostecki wrote: > On Wed, Feb 10, 2021 at 05:24:28AM +0100, Michał Mirosław wrote: > > On Tue, Feb 09, 2021 at 09:30:40PM +0100, Michal Rostecki wrote: > > [...] > > > For the array with 3 HDDs, not adding any penalty resulted in 409MiB/s > > > (429M

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-10 Thread Michal Rostecki
On Wed, Feb 10, 2021 at 05:24:28AM +0100, Michał Mirosław wrote: > On Tue, Feb 09, 2021 at 09:30:40PM +0100, Michal Rostecki wrote: > [...] > > For the array with 3 HDDs, not adding any penalty resulted in 409MiB/s > > (429MB/s) performance. Adding the penalty value 1 resulted in a > > performance

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-10 Thread Anand Jain
On 10/02/2021 04:30, Michal Rostecki wrote: From: Michal Rostecki Add a new raid1 read policy `roundrobin`. For each read request, it selects the mirror which has lower load than queue depth and it starts iterating from the last used mirror (by the current CPU). Load is defined as the number of

Re: [PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-09 Thread Michał Mirosław
On Tue, Feb 09, 2021 at 09:30:40PM +0100, Michal Rostecki wrote: [...] > For the array with 3 HDDs, not adding any penalty resulted in 409MiB/s > (429MB/s) performance. Adding the penalty value 1 resulted in a > performance drop to 404MiB/s (424MB/s). Increasing the value towards 10 > was making th

[PATCH RFC 6/6] btrfs: Add roundrobin raid1 read policy

2021-02-09 Thread Michal Rostecki
From: Michal Rostecki Add a new raid1 read policy `roundrobin`. For each read request, it selects the mirror which has lower load than queue depth and it starts iterating from the last used mirror (by the current CPU). Load is defined as the number of inflight requests + a potential penalty value