We're hiring!
*

NVMe: Officially faster for emulated controllers!

Helen Koike avatar

Helen Koike
June 13, 2017

Share this post:

The Doorbell Buffer Config command

When I last wrote about NVMe, the feature to improve NVMe performance over emulated environments was just a living discussion and a work in progress patch. However, it has now been officially released in the NVMe Specification Revision 1.3 under the name "Doorbell Buffer Config command", along with an implementation that is already in the mainline Linux kernel! \o/

You can already feel the difference in performance if you compile Kernel 4.12-rc1 (or later) and run it over a virtual machine hosted on Google Compute Engine. Google actually updated their hypervisor as soon as the feature was ratified by the NVMe working group, even before it was publicly released.

There were very few changes from the original proposal, I.e. opcodes, return values and now fancy names; the buffers (as described in my last post) are now called Shadow Doorbell and EventIdx buffers.

In short, the first one mimics the Doorbell registers in memory, allowing the emulated controller to fetch the Doorbell value when convenient instead of waiting for the Doorbell register to be written. For its part, the EventIdx provides a hint given by the emulated controller to tell the host if the Doorbell register needs to be updated (in case the emulated controller is not fetching the Doorbell value from the Shadow Doorbell buffer). You can check section 7.13 of the specification for an example of usage.

Results

The following test results were obtained in a machine of type n1-standard-4 (4 vCPUs, 15 GB memory) at Google Cloud Engine platform with Kernel 4.12.0-rc5 using the following command:

$ sudo fio --time_based --name=benchmark --runtime=30 \ --filename=/dev/nvme0n1 --nrfiles=1 --ioengine=libaio --iodepth=32 \ --direct=1 --invalidate=1 --verify=0 --verify_fatal=0 --numjobs=1 \ --rw=randread --blocksize=4k --randrepeat=0

Results (in Input/Ouput Operations per Second):
Without Shadow Doorbell and EventIdx buffers: 43.9K IOPS
With Shadow Doorbell and EventIdx buffers: 184K IOPS
Gain ~= 4 times

Screenshot - Without Shadow Doorbell and EventIdx buffers


Screenshot - With Shadow Doorbell and EventIdx buffers


Enjoy your enhanced numbers of IOPS! :D

 

Original post

Comments (0)


Add a Comment






Allowed tags: <b><i><br>Add a new comment:


Latest Blog Posts

Permissively-licensed MTP device implementation

16/05/2019

Introducing cmtp-responder - a permissively licensed Media Transfer Protocol (MTP) responder implementation which allows embedded devices…

An eBPF overview, part 5: Tracing user processes

14/05/2019

Up until now, talking in-depth about userspace tracing was deliberately avoided because it merits special treatment, hence this part devoted…

CEF on Wayland upstreamed

08/05/2019

After a successful team effort, the patch enabling the Chromium Embedded Framework (CEF) Ozone builds to run with different platform backends,…

An eBPF overview, part 4: Working with embedded systems

06/05/2019

Now that we've studied the mainstream way of developing and using eBPF programs on top of the low-level VM mechanisms, we'll look at projects…

Running Android and Wayland on embedded devices

02/05/2019

A previous post introduced the SPURV Android compatibility layer for Wayland based Linux environment. In this post, we're going to dig into…

An eBPF overview, part 3: Walking up the software stack

26/04/2019

In part 1 and 2 of this series, we took a condensed in-depth look at the eBPF VM. In part 3, we define the high-level components of an eBPF…

Open Since 2005 logo

We use cookies on this website to ensure that you get the best experience. By continuing to use this website you are consenting to the use of these cookies. To find out more please follow this link.

Collabora Ltd © 2005-2019. All rights reserved. Website sitemap.