References: <91c96949-ec5c-04f2-b198-368f1e04e957@redhat.com> <88bfd65fc15f59c8781ab3f626811f79@assyoma.it> From: Zdenek Kabelac Message-ID: <0d0d022b-93ea-156d-c408-49efbc31bbb2@redhat.com> Date: Mon, 20 Jan 2020 15:40:57 +0100 MIME-Version: 1.0 In-Reply-To: <88bfd65fc15f59c8781ab3f626811f79@assyoma.it> Content-Language: en-US Content-Transfer-Encoding: quoted-printable Subject: Re: [linux-lvm] Time needed for take a snapshot Reply-To: LVM general discussion and development List-Id: LVM general discussion and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , List-Id: Content-Type: text/plain; charset="iso-8859-1"; format="flowed" To: LVM general discussion and development , Gionatan Danti Dne 20. 01. 20 v 11:36 Gionatan Danti napsal(a): > Il 20-01-2020 10:22 Zdenek Kabelac ha scritto: >> So having=EF=BF=BD thousands of LVs in a single VG will become probably = your=20 >> bottleneck. >=20 > Hi Zdenek, I was thinking more about having few LVs, but with different a= mount=20 > of data/mapping. >=20 > For example, is a very fragmented volume (ie: one written randomically)=20 > significantly slower to snapshot than an almost empty volume? I fully exp= ect=20 > some small difference; however, if an empty volume take 0.2s and a fragme= nted=20 > one 20s, this would surely be significant. >=20 > Note that I never had such a slow snapshot, rather, even on aged and big = > volumes, it always take <1s. However, other experiences are welcome. Yep - kernel metadata 'per thin LV' are reasonably small - so even for big = thin devices it still should fit within your time boundaries. (effectively thin snapshot just increases 'mapping' sharing between origin = and=20 its snapshot - so the time needed depends on how many bTree nodes needs to = be=20 updated - so if you would manage to create heavily fragmented multi TiB=20 thinLV, the time depends on speed of you metadata device - as long as devic= e=20 is fast (i.e. >=3D SSD) - operation should be quick. But ATM there is no scientific proof for the worst case scenario. Regards Zdenek