Supported Ceph from SUSE – when you need one less enterprise plate to keep spinning.

Running an enterprise IT environment at scale can be a tricky business – a bit like the circus performer’s spinning plates. Everyone has a few of these: processes or platforms that need a little bit of a helping hand to stay up, and everyone has the occasional slip. You know how it is: as soon as you’ve got one thing working perfectly, another plate slows its rotation, starts to teeter, and if you’re not quick you will be looking at shards of broken crockery.

Keeping those plates spinning is easier for the circus performer than it is for the IT team. In the big tent, all the plates are the same size and weight, the rods supporting them are the same height, and they are all in the same physical geographic location, with a convenient and uniform distance between them.

When a plate hits the floor in the big tent it might not ‘wow’ the audience, but at least it doesn’t affect the other acts: the clowns continue chuck buckets of water at each other, the trapeze artists continue to fly through the air, the acrobats to tumble. . . . when your storage goes down, applications and business processes go with it. The punters in big tent might not be too impressed – but hey it was impressive to watch 20 plates at the same time and something had to go sooner or later, right?

Not so for IT. The plates – applications, servers, business processes – are all different sizes and weights the rods are different heights – they are even in different places (sometimes not even in the same country). And rather than just one guy racing up and down a neat row dealing a deft and judicious flick of the wrist to keep things in motion, you’ve got whole teams.

And those applications are not of equal importance. Some go down and the business barely notices. You can take your time, get the system back up again. Others are important enough to merit a service level agreement. When things go wrong with these more critical services your ‘audience’ AKA really important people – don’t give a damn about the other 20 plates you’ve still got spinning: the only one they ever noticed or cared about was their own. And you broke it. That’s the trouble with IT firefighting – everyone loves it when you put the fire out. . . . but somehow they think it was you that set the house in fire in the first place.

Here at SUSE we hear stories on a daily basis about organizations who have been early adopters of Ceph. Faced with ever growing costs, increasing volumes, complexity and driven by fear of vendor lockdown, they look to open source software defined storage as the answer. They know the cost savings are there, they know that the route to the software defined data centre lies this way, and they know they are going to need to change their skillsets for a digitally transformed future where the nature of the IT team and the roles it performs have changed. Some rush in a little too fast – trying to get ahead of the adoption curve, and for a few – it works out as a jump from the frying pan into the fire.

So, if you’ve gone hot-foot into Ceph for all the right reasons but burnt your feet in the process, maybe its time to stop firefighting and call in a crew that prevent the fires from breaking out in the first place. There’s plenty of clear advantage in managed support:

1. Hand over the headaches – the SLA becomes someone else’s problem, minimize downtime, and eliminate maintenance hassle.
2. Stay ahead of the adoption curve: open source projects deliver innovation at breakneck speed – that translates into a LOT of software upgrade grief which expert service eliminates.
3. Have just one place to call when trouble occurs
4. Benefit from the experiences of others
5. Get exclusive value adds.

So, if your mission to modernize has left you looking at broken IT crockery a little too often and your business stakeholder audience is a little prone to cat-calling and likes to ‘boo’, maybe its time you called the team at SUSE.