Survey Says! (Part One)
As a reflection of growing leadership in the distributed DBMS market, NuoDB has been the subject of a lot of research lately.
In just the past month, the company and/or the NuoDB product have been cited in:
- the first-ever Gartner Magic Quadrant for Operational DBMS;
- a 451 Research Impact Report by Matt Aslett, a leading database analyst; (also available here).
- a thorough analysis by Phil Howard of Bloor Research; Phil has covered the data management market for 40 years;
- the third quarter Cloud round up by GigaOM Research.
This doesn’t include opinions in over a dozen blogs and articles. From Enrico Signoretti in his widely-read blog “Juku,” to Jack Clark in The Register, to ODBMS by Roberto Zicari, there have been all kinds of conjecture about NuoDB.
With so much data around even we have had a hard time sorting it all out. But that’s my job so I took it on.
When you aggregate up what all the gurus have to say, I boil it down to 3 key trends:
- The DBMS market is consolidating. For example, out of dozens and dozens of startup DBMS companies, Gartner stripped the competitive landscape down to just 19 players to watch in its first ever Operational DBMS Magic Quadrant.It put all of them in or (very close to) the niche vendor quadrant. Only the four traditional vendors (Oracle, SAP, IBM and Microsoft) are slotted into the coveted upper right quadrant – the leaders corner. If Oracle’s falling quarterly financial reports are any indication, this quadrant is the land of 30-year-old technology that is steadily being supplanted by upstarts hungering for motion up and to the right.Gartner typically updates their MQ analyses every 12 months however they will refresh this first-time-ever MQ next June. You can see the pattern. Lots of movement, consolidation underway, changes to come.
- It’s not your grandfather’s OLTP any more. Call it what you will (NewSQL is our favorite; 451’s Matt Aslett invented the term; Phil Howard also refers to NewSQL; Gartner calls it operational DBMS; Enrico Signoretti just says “sexy!”), handling transactions is simply not enough any more.New requirements – from real time operational intelligence – to the ability to handle new apps (think smartphones, think social media) – to an absolute requirement for both elastic scalability and continuous availability – are all part of a mandatory switch from pure transaction processing to operational DBMS.
- Don’t despair… distribute! OK, it’s self-serving to say deploying a single, logical database in multiple locations at the same time is an essential requirement for DBMS in 2014. Self-serving because NuoDB is the only solution that can do this with SQL/ACID.But, never mind, when you take a look at what the analysts and pundits are saying, they may use different words but they mean gotta have “geo-distribution.”Here’s why. A truly distributed DBMS with ACID transactional guarantees addresses the pain points I mentioned above. Among these the most notable are on-demand capacity, continuous availability and geo-distributed operation also providing low latency and automatic redundancy. Traditional designs unfortunately deliver much less than those promised advantages and involve costs, complexities and/or functional limitations that limit their usefulness and general applicability. And, some of the newer solutions lack the familiarity of SQL (no retraining required) and ACID guarantees.
So, this blog post skims the surface of all this new analysis and opinions.
However, to be sure we got the market trends right from the customers’ perspective, we hired the largest independent research company in the world – IDG – to survey a large number of IT management who are not NuoDB customers.
My next blog will cover their responses to the IDG survey.