xfs
[Top] [All Lists]

Re: Bad Metadata performances for XFS?

To: Chris Murphy <lists@xxxxxxxxxxxxxxxxx>
Subject: Re: Bad Metadata performances for XFS?
From: Roger Willcocks <roger@xxxxxxxxxxxxxxxx>
Date: Wed, 06 Jul 2016 12:49:29 +0100
Cc: Wang Shilong <wshilong@xxxxxxx>, "linux-xfs@xxxxxxxxxxxxxxx" <linux-xfs@xxxxxxxxxxxxxxx>, "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <CAJCQCtRiznrWAmCEgd=uMETnBubh1ZttGP4xF9PLu0_j7XBRuw@xxxxxxxxxxxxxx>
References: <3ED34739A4E85E4F894367D57617CDEF9ED9518B@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> <20160704225226.GD27480@dastard> <20160705001854.GY12670@dastard> <3ED34739A4E85E4F894367D57617CDEF9ED955AB@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> <CAJCQCtRiznrWAmCEgd=uMETnBubh1ZttGP4xF9PLu0_j7XBRuw@xxxxxxxxxxxxxx>
On Tue, 2016-07-05 at 14:34 -0600, Chris Murphy wrote:
> On Mon, Jul 4, 2016 at 7:43 PM, Wang Shilong <wshilong@xxxxxxx> wrote:
> >
> >
> > I understand that this is single thread Limit, but I guess there are some
> > other Limit here, because even single thread creating 50W files speed
> > is twice than 200W files.
> 
> Watts or Wolframs (tungsten)?
> 
> 50W!=50000. You could write it as 50k and 200k. It's unlikely to get
> confused with 50K and 200K, which are temperatures, because of
> context. But W makes no sense.
> 
> 
> 

I suspect it's an abbreviation for the (Chinese) unit 'wan'

https://www.quora.com/Why-do-Chinese-people-count-in-units-of-10-000

so it makes perfect sense but it's not an SI unit.

--
Roger


<Prev in Thread] Current Thread [Next in Thread>