Come on everyone does it! It’s the last day of 2015, so of course I’m going to write a blog to highlight 3 trends I’ve seen in planet Database this year. Think yourself lucky that I am not dusting off the crystal ball to give you my amazing, insightful predictions for 2016…… like a potential new IoT analytics platform for farmers called Pig Data…. (Ok.. I’ll stop now, before I get started).
I’m not an analyst or a fortune teller, so I’m going to play it safe and stick to listing a few things that I have seen happening with our customers.
Oracle are struggling to win customer’s database hearts and minds (esp. mid-Market)
OK this is an obvious one I hear you cry and I think the evidence is there to support that Oracle may not really be trying that hard.
- Licensing policy on VMware or indeed most cloud or virtualisation platforms
- Frequency of customer audits
- The latest SE2 changes, lowering the CPU count from SE
- The lack of a Europe based “useable” cloud offering (don’t get me started on this one!)… etc…
What is absolutely happening in the customers we have spoken to is a couple of things. Firstly more and more Oracle ISV’s are adding SQL Server as an alternative/replacement meaning that new application deployments using Oracle databases are going to diminish. Secondly, Microsoft are mopping up as customers move systems onto virtual and cloud platforms. I get the impression that Oracle’s interests lie in the full Red Stack strategy for large enterprises and SaaS delivered Oracle owned applications, happy to be proved wrong in 2016.
NoSQL has its place, but vendors are under pressure.
We’ve spent a lot of time in 2015 working with NoSQL. What we’ve found is that for the right use case NoSQL is a very good fit, but I’m not convinced the addressable market many of the vendors would have you believe is actually there. To understand the sweet spots, you really need to go back to what NoSQL is and that is largely a simple database that is designed for speed and handling data that lacks structure. So it’s perfect for projects such as personalisation, profile management, IoT, product catalogues etc.. but it’s not going to replace your corporate, application dependant Oracle database any time soon.
The big problem moving forward is that many of the NoSQL vendors in this competitive market have received mega investment and to realise this they need to show a larger and larger addressable market. You just have to look at the product development and roadmaps to see they are adding features to appeal to the masses. It’s certainly a space to watch, I just hope they don’t sell out their technology roots to try and show an investor they have the features to compete with Oracle DB & Microsoft SQL Server.
Customer’s maturity needs outgrowing Database as a Service platforms
First to qualify, by database as a service I mean the automated, shrink wrapped, DBaaS software platforms such as Amazon RDS, Compose.io, Object Rocket etc.. To a certain extent this trend has surprised me a little and to be honest I’m not sure if it’s the tip of a trend to emerge in 2016 or just we have had a run of similar customer projects.
The main trend we have seen is customers who have started developing using a DBaaS offering and then they have hit one of the below walls, mainly because the project is growing and under more scrutiny:
- An outage or service issue which has amplified the fact that they don’t have access to a team of DBA’s at their beckon call to fix the issue. Customers like the autonomy of the service, but when something goes wrong they are missing the personable DBA support.
- They have hit a technical constraint. Most of these platforms offer vanilla, one:many packaged installations, so when a project needs something like SSIS for example and they are running RDS, then they are forced to look at alternatives.
- They need deeper dive access into the database. One example is performance tuning, an experienced DBA just doesn’t have access to tune the same things they could on a full blown database vendor install.
It think this really represents the maturity of customers environments, they start with an easy solution, if things go well, it gets more serious. The solution at least for us has been an easy one and typically we deliver an agile managed service on the chosen cloud platform, AWS being by far the most popular.
That pretty much wraps up my 2015 ramblings, have a great New Year!