The first thing you generally want to do when you have any new Storage system like SSD, Disk arrays or a Cluster Ceph, is benching. You will want to know how can read and write throughput. FIO is able to do that for you, here is an example:
[global] ioengine=libaio invalidate=1 ramp_time=5 direct=1 size=5G runtime=300 time_based directory=/home [seq-read] rw=read bs=64K stonewall [rand-read] rw=randread bs=4K stonewall [seq-write] rw=write bs=64K stonewall [rand-write] rw=randwrite bs=4K stonewall You then will have a good output of everything you need to know.
Do not miss OpenStack meetup event at eNovance headquarter! There will be big talks on:
ZFS Ceph Swift You’ve understood, it’ storage related for OpenStack.
Of course I’ll be there :-D. More informations are available here.
This is a very good news if you’re using Ceph in production! Originally delivered as a proprietary dashboard included with Inktank Ceph Enterprise, Calamari has some really great visualization stuff for your cluster as well as the long term goal of being the all-in-wonder management system that can configure and analyze a Ceph cluster.
Calamari is composed in 2 elements:
Backend: the Calamari backend is written in Python 2.6+, using Saltstack, ZeroRPC, gevent, Django, django-rest-framework, graphite, (and a few others I may have forgotten) and instantiates a new REST API for integration with other systems.
In France, we have the chance to have an ISP/Mobile Telephony provider called Free. They considerably reduced the market price for the Mobile phone communications in 2012. This was very fine for competition and for French people. Now we can have correct prices for Internet and cell phone data. Thanks to Free.
Ceph is quite the same as Free but for storage systems. Since many years, disks array constructors sold us locked solutions with a very high cost of hardware and licenses’ software.
One of my colleague (Sebastien Han) started to play (like me) with Ansible. As he is a Ceph master, he often deploy Ceph for tests or production. He generally uses Puppet for that purpose.
But due to some Puppet limitations or big brain fuck it involves, he started to play with Ansible and created a Ceph module for it. He showed me it, when he was writing it and it seamed working like a charm.
I’ll be once again happy to participate to Ceph Day ! We’ll see the new version of Ceph, best practices and how to scale it right :-)
This sounds good ! To get more informations: https://www.eventbrite.com/e/ceph-day-frankfurt-tickets-10173269523
Thanks for eNovance to send me there, I’m really happy !
Don’t miss the Ceph Meetup on Monday ! You will be aware of the new features in Firefly (like tiering or erasure coding) ! And this is the first long time support release !!!
To get more informations: http://www.meetup.com/Ceph-in-Paris/events/158942372/