Virtualization Technology News and Information
Article
RSS
NuoDB 2016 Prediction: The On-demand Economy is Driving Database Convergence and Vendor Consolidation

Virtualization and Cloud executives share their predictions for 2016.  Read them in this VMblog.com series exclusive.

Contributed by Barry Morris, co-founder and executive chairman, NuoDB

The On-demand Economy is Driving Database Convergence and Vendor Consolidation

It's no secret that we are transitioning to an on-demand economy -- a world where we expect things to happen ‘here and now' as soon as our smartphones issue the imperatives. That's good news for impatient consumers and very bad news for database systems.

The database industry has seen three decades of gradual evolution, followed by a few years of mold-breaking innovation, driven by on-demand needs and datacenter trends.  The many themes of that innovation include Scale-out, Big Data, NoSQL, NewSQL, Stream Processing, DBaaS, In-memory, Cloud and more. Along with these themes, we've seen a proliferation of companies, technologies and open source projects. 

The current generation database landscape is still dominated by a few large and traditional players, while the next generation landscape has become highly fragmented and confusing.  The laws of software economics tell us that this too shall pass.  We have seen it before. The next few years will witness technology convergence and market consolidation. 

So if the on-demand economy is driving a cycle of database innovation and re-consolidation, then what are the big trends, and predictions for 2016? Let me share my opinion with you.

1.  Multi-model

We are moving into a post-NoSQL world, a world that is both Not-only-SQL and Not-only-NoSQL.  Expect your database vendor to offer all-of-the-above as they are all moving towards a multi-model offering. 

Most SQL databases support or will soon support JSON, including ORACLE 12.1.0.2, SQL Server 2016 and IBM DB2 10.5. Many also support graph/RDF. Simultaneously NoSQL vendors are adding limited transactions and SQL. Furthermore the Hadoop ecosystem has moved from Map/Reduce to SQL, with at least 10 separate SQL offerings, including Hive, Impala and HP Vertica.

Never mind the details.  The point is that customers won't run two or three databases per application. The future is multi-model.

2. Local Everywhere

One of today's buzzwords is the Internet of Things. The reality according to Gartner is that by 2020, 25 billion devices will be generating data about almost every topic imaginable. And we'll be collecting transactional data from toothbrushes to wineglasses, daily. These connected devices are scattered across the globe, everywhere and users will expect instant reactions to their actions.  It's what technical folks call low-latency response.  Mobile and web applications are no different. 

What is needed is local access to global data management systems, preferably with a "single global truth".  Databases can't be single-server systems running in a single datacenter.  They must be distributed, and in fact geographically distributed in order to be local everywhere.  This is a big 2016 topic as IoT moves from excitement over devices to head scratching over managing the data they generate.

3. Memory-first

A server with one terabyte of memory (DRAM) now costs around $20,000.  Memory capacity per dollar is doubling every 18 months.  We should expect that by the end of 2016 commodity servers will offer close to one terabyte of memory, and therefore that you should be thinking about a database system that is designed to be "memory-first".  Your valuable data will be permanently in memory with backing store, not stored on disk and occasionally loaded into memory.  Memory-first is an upside-down architecture compared to the old way of doing things.  The traditional RDBMS products are fantastically powerful, but they will be less and less able to exploit new age hardware including "free" memory and low cost solid-state storage.

I expect the in-memory revolution to become much more mainstream in 2016, as CIOs find that all but the most data-hungry applications are no longer limited by slow disk reads. The database world is gearing up to deliver on this situation.

4. Data Containers

You say you want a revolution? Containers, such as Docker, are legitimately an economic disruptor in the datacenter. At one level they are lighter weight virtual machines -- which sounds like a nice-to-have. But at another level containers are much more agile and better suited for dynamic load balancing, for supporting application models -- micro-services in particular -- and enabling lower cost datacenter management at scale. You might not be moving your legacy applications to Docker any time soon, but your competitors will be building new and equivalent applications that are suited to container deployment and operating at higher velocity with much lower costs.  

Containers are not business as usual in the database world.  Using traditional database systems in the world of Containers is like inventing the airplane and then welding it to a battleship - in both cases you negate the benefit of the innovation.  Application containers need an equally virtualized data management layer.  Let's call it a data container layer.  

In 2016 you will be hearing a lot more about containers, especially Docker, and how Docker is substantially less Docker-esque in the absence of database systems that support container architectures. You will also see most SQL and NoSQL database vendors wringing their hands about this being a hard problem to solve, while a few others will be offering products that are designed to be data containers.

5. Data Residency

Lets shift to another issue - data residency -- that's making headlines, especially in Europe. As we know, several European and Asian countries have privacy legislation demanding that local data stays in their respective countries. That sounds all good and fair. But the implication is far-reaching and hard to solve. Even large companies like Google and Microsoft are not prepared to deal with such regulations without violating local laws. According to a survey, more than 70% of cloud providers do not comply with European data privacy legislation.

Granted it's a hard problem to solve. However, companies such as NuoDB has a solution for that. More importantly, we are working with a standards body and leading a Data Residency Working Group to solve this challenge for large and small organizations.

##

About the Author
Barry Morris, co-founder and executive chairman, NuoDB

As executive chairman, Barry chairs the Board of Directors and drives the external matters of NuoDB, including industry thought leadership, strategic partnerships and business relationships as well as product and technology strategy. He also serves as an advisor to the CEO and senior leadership on all aspects of the business.

Barry is an accomplished software executive with more than 25 years of industry experience in the U.S. and Europe -- startups to publically held companies. As the founding CEO of NuoDB, Barry built the company from ground up into a successful database company and a leader in scale-out SQL technology.

Prior to founding NuoDB, Barry was the CEO at IONA Technologies, where he led Ireland's most successful software company through its strongest period of growth, with more than $180 million in revenues. He was subsequently chairman and chief executive officer of StreamBase Systems, where he turned the company into the market leader in Complex Event Processing (CEP). Barry has also held leadership roles in various companies in the U.S. and Europe, including PROTEK, Metrica, Lotus Development and DEC.

Born in South Africa, Barry lived in England and Ireland before moving to the Greater Boston Area. He holds a bachelor's degree in engineering from New College Oxford University and an honorary doctorate in Business Administration (DBA) from the IMCA. He also serves on boards of several startups in Boston, Ireland and South Africa.

Published Tuesday, October 20, 2015 6:29 AM by David Marshall
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
top25
Calendar
<October 2015>
SuMoTuWeThFrSa
27282930123
45678910
11121314151617
18192021222324
25262728293031
1234567