On Fri, Mar 30, 2018 at 09:04:46AM +, Eric H. Chang wrote:
> We internally call PCIe-retimer as HBA. It's not a real Host Bus Adapter that
> translates the interface from PCIe to SATA or SAS. Sorry for the confusion.
Please don't call a PCIe retimer an "HBA"! :)
While your experiment is setu
@@ -85,7 +121,14 @@ static blk_qc_t nvme_ns_head_make_request(struct
request_queue *q,
int srcu_idx;
srcu_idx = srcu_read_lock(&head->srcu);
- ns = nvme_find_path(head);
+ switch (head->mpath_policy) {
+ case NVME_MPATH_ROUND_ROBIN:
+ ns = nvme_fin
For PCIe devices the right policy is not a round robin but to use
the pcie device closer to the node. I did a prototype for that
long ago and the concept can work. Can you look into that and
also make that policy used automatically for PCIe devices?
I think that active/active makes sense for
On Fri, Mar 30, 2018 at 01:57:25PM +0900, Baegjae Sung wrote:
> Our prototype uses dual-ported PCIe NVMe connected to a single host. The
> host's HBA is connected to two switches,
What "HBA"? We are talking about NVMe here..
2018-03-29 4:47 GMT+09:00 Keith Busch :
> On Wed, Mar 28, 2018 at 10:06:46AM +0200, Christoph Hellwig wrote:
>> For PCIe devices the right policy is not a round robin but to use
>> the pcie device closer to the node. I did a prototype for that
>> long ago and the concept can work. Can you look in
On Wed, Mar 28, 2018 at 01:47:41PM -0600, Keith Busch wrote:
> single host (single path from two hosts seems more common). If that's a
> thing, we should get some numa awareness. I couldn't find your prototype,
> though. I had one stashed locally from a while back and hope it resembles
> what you h
On Wed, Mar 28, 2018 at 10:06:46AM +0200, Christoph Hellwig wrote:
> For PCIe devices the right policy is not a round robin but to use
> the pcie device closer to the node. I did a prototype for that
> long ago and the concept can work. Can you look into that and
> also make that policy used auto
For PCIe devices the right policy is not a round robin but to use
the pcie device closer to the node. I did a prototype for that
long ago and the concept can work. Can you look into that and
also make that policy used automatically for PCIe devices?
Some storage environments (e.g., dual-port NVMe SSD) provide higher
performance when using multiple paths simultaneously. Choosing a
path from multiple paths in a round-robin fashion is a simple and
efficient way to meet these requirements.
We implement the active-active round-robin path selector
9 matches
Mail list logo
