Hmm change 40Gbps to 100Gbps networking. 40Gbps technology its just a bond of 4x10 Links with some latency due link aggregation. 100 Gbps and 25Gbps got less latency and Good performance. In ceph a 50% of the latency comes from Network commits and the other 50% from disk commits. A fast graph : https://blog.mellanox.com/wp-content/uploads/John-Kim-030416-Fig-3a-1024x74… Article: https://blog.mellanox.com/2016/03/25-is-the-new-10-50-is-the-new-40-100-is-… Micron got their own Whitepaper for CEPH and looks like performs fine. https://www.micron.com/-/media/client/global/documents/products/other-docum… AS your Budget is high, please buy 3 x 1.5K $ nodes for your monitors and you Will sleep better. They just need 4 cores / 16GB RAM and 2x128GB SSD or NVME M2 . -----Mensaje original----- De: Adam Boyhan <adamb(a)medent.com> Enviado el: viernes, 31 de enero de 2020 13:59 Para: ceph-users <ceph-users(a)ceph.io> Asunto: [ceph-users] Micron SSD/Basic Config Looking to role out a all flash Ceph cluster. Wanted to see if anyone else was using Micron drives along with some basic input on my design so far? Basic Config Ceph OSD Nodes 8x Supermicro A+ Server 2113S-WTRT - AMD EPYC 7601 32 Core 2.2Ghz - 256G Ram - AOC-S3008L-L8e HBA - 10GB SFP+ for client network - 40GB QSFP+ for ceph cluster network OSD 10x Micron 5300 PRO 7.68TB in each ceph node - 80 total drives across the 8 nodes WAL/DB 5x Micron 7300 MAX NVMe 800GB per Ceph Node - Plan on dedicating 1 for each 2 OSD's Still thinking out a external monitor node as I have a lot of options, but this is a pretty good start. Open to suggestions as well! _______________________________________________ ceph-users mailing list -- ceph-users(a)ceph.io To unsubscribe send an email to ceph-users-leave(a)ceph.io

The use case is for KVM RBD volumes. Our enviroment will be 80% random reads/writes probably 40/60 or 30/70 is a good estimate. All 4k-8k IO sizes. We currently run on a Nimble Hybrid array which runs in the 5k-15k IOPS range with some spikes up to 20-25k IOPS (Capable of 100k iops per Nimble). Its also worth mentioning that I plan on splitting this cluster between our two sites. We have a dedicated dark fiber connection which is 8x 10GB links between the two sites (roughly .4 or .5 ms latency). I was leaning towards 40G because our existing HP 5900 switch's which we use to light our dark fiber also have 4x 40GB ports which I could use to hang the Ceph cluster network from. This would be a 4/2 setup, keeping 2 copies at each site. I currently have a much lower end ceph cluster using all spinning rust with 3 nodes at each site (48 OSD's). So far that is working out really well, but the data going on that cluster is written than almost never touched again. The above spinning rust cluster is running just simple 10G SFP+.