What happened to SQL-NoSQL Campwars?

Pratima H

BANGALORE, INDIA: The database world was a boring but a calm place when just a few names from the Relational Royalty and SQL folks ruled the space. Things however started to shake with the influx of open source RDBMSs (Relational Database Management Systems), the onslaught of MySQL, Oracle’s twist jab with the acquisition of Sun Microsystems and hence MySQL, the ensuing ‘save MySQL’ protest-waves, the entry of Big Data and modern-workloads and with them the new kid on the block- NoSQL.

Today, names like Cassandra, Hadoop, UnSQL, CQL, in-memory engines and what not are said in the same breath as other marquee names from yesteryear’s database roster.

Some industry watchers have been foreboding death of the heavy relational tables for quite some time, and the rest have been caught between ‘SQL does not pack the muscle for Big Data scenarios’ and ‘NoSQL is still not mature, robust and scalable beyond in-memory’ debates.

Amidst this fierce SQL-NoSQL divide, Gartner indicates in its latest report titled ‘The State of Open Source RDBMSs, 2015 that OSRDBMSs(Open Source Relational Database Management Systems ) have matured to the point where they can replace commercial databases.

Can CIOs really consider them as a standard infrastructure choice, as the research firm recommends? What is happening to proprietary RDMBMs if OSRDBMSs are indeed being used by more than 25 per cent of the market with the entire open source DB market (including relational and non-relational) hitting $562m in revenue last year (yes, at a 31 per cent clip over 2013 vs the 5.4 per cent in the preceding year)?

Who perches where when it comes to equally important but sometimes conflicting dimensions like analytics, scale, application agility, developer ease and excitement?

On the margins of the SQL and Data Conference, SQLServerGeeks Annual Summit 2015 we caught up with Denny Cherry, Principal Consultant, DC&A, USA. As the owner and principal consultant for Denny Cherry & Associates Consulting and with a decade of experience working with platforms such as Microsoft SQL Server, Hyper-V, vSphere and Enterprise Storage solutions and expertise around system architecture, performance tuning, security, replication and troubleshooting; Denny shares his unique perspectives on the current contours of Database industry landmines.

What, in your opinion, is a recent highlight that has defined this space strongly?

I would say SQL Server 2016 is a good turn in terms of high availability, increased throughput and large workloads. Microsoft has brought in bug functionality changes and it is claiming this as the biggest leap forward in Microsoft's data platform history with real-time operational analytics, rich visualizations on mobile devices, built-in advanced analytics, new advanced security technology, and new hybrid cloud scenarios. It is also hoped that it would be apt for mission-critical capabilities with in-memory performance and operational analytics built-in.

What about the good old SQL-NoSQL split? Has the divide been shrinking?

Both relational and non-relational products have their users. Neither is better or vice versa. Some things can and should be done with SQL (like transactions at banks where consistency counts) and for other new kind of workloads, NoSQL can be apt. Both are gaining complimentary shape, is how I see it.

Where and how does it impact the big boys like Oracle and Microsoft?

They have started to figure out changes prudently. Their portfolios, like that of Microsoft, are reflecting NoSQL direction.

Any ripples worth noting on the skill-side?

Today, a lot of stuff is going to Cloud than what it was two years back. People will have to learn and keep up else they will be left behind. Cloud is not that scary either and it is not going to take jobs away.

What challenges confront SQL side though?

People need to make time to learn new features. If IT professionals stop learning, then the whole curve breaks.

Can SQL now stand up for Big Data challenge?

Yes, it can handle those workloads. Instances on Microsoft’s Azure platform are a good example of how tons of rows can be handled well.

What challenges or gaps confront the NoSQL side? Is it still unstructured?

Every workload has a different learning curve and same goes with unstructured ones. People need to figure out what really fits their problem instead of being scared of confusion or new stuff.

Where would in-memory names like HANA, Exadata etc affect the space, if at all?

Yes, these engines have moved from concepts to phenomenal features. At the same time there are lot of advancements in SQL Server too where it is not about a separate engine but a situation where people can choose what piece to opt for an in-memory functionality. New versions will have some impressive progress.

SQL injections and other hacks – how vulnerable does the security side stay? Has the innate network element and connectedness of Cloud environments made the challenge harder?

Security, in general is turning in good form. Environments in Cloud platform are also doing well. They have good firewalls and most break-ins happen due to two main reasons – social engineering and poor application code.

How is the dance with the Open Source world going?

It is definitely a complimentary scenario and communities like SQL Server are amplifying well the idea of sharing information where and as fast as it matters and as wide as possible.