Databases Reference
In-Depth Information
12.7
Summary
In this chapter, we've looked at how a formal architecture trade-off process can be
used to select the right database architecture for a specific business project. When
architectures were few in number, the process was simple and could be done infor-
mally by a group of in-house experts. There was no need for a detailed explanation of
their decisions. But as the number of NoSQL database options increases, the selection
process becomes more complex. You need an objective ranking system that helps you
narrow down the options and then compares the trade-offs.
After reading the cases studies, we believe with certainty that the NoSQL move-
ment has and will continue to trigger dramatic cost reductions in building applica-
tions. But the number of new options makes the process of objectively selecting the
right database architecture more difficult. We hope that this topic helps guide teams
through this important but sometimes complex process and helps save both time and
money, increasing your ability to adapt to changing business conditions.
When Charles Darwin visited the Galapagos Islands, he collected different types
of birds from many of the islands. After returning to England, he discovered that a
single finch had evolved into roughly 15 different species. He noted that the size and
shape of the birds' beaks had changed to allow the birds to feed on seeds, cacti, or
insects. Each island had different conditions, and in time the birds evolved to fit the
requirements.
Seeing this gradation and diversity of structure in one small, intimately
related group of birds, one might really fancy that from an original paucity of
birds in this archipelago, one species had been taken and modified for differ-
ent ends. (Charles Darwin,
Voyage of the Beagle
)
NoSQL databases are like Darwin's finches. New species of NoSQL databases that
match the conditions of different types of data continue to evolve. Companies that try
to use a single database to process different types of data will in time go the way of the
dinosaur. Your task is to match the right types of data to the right NoSQL solutions. If
you do this well, you can build organizations that are healthy, insightful, and agile,
and that can take advantage of a changing business climate.
For all the goodness that diversity brings, standards are still a must. Standards allow
you to reuse tools and training and to leverage prebuilt and preexisting solutions.
Metcalf's Law, where the value of a standard grows exponentially as the number of
users increases, applies to NoSQL as well as to network protocols. The diversity-
standardization dilemma won't go away; it'll continue to play a role in databases for
decades to come.
When we write reports for organizations considering NoSQL pilot projects, we
imagine Charles Darwin sitting on one side and Robert Metcalf sitting on the other—
two insightful individuals using the underlying patterns in our world to help organiza-
tions make the right decision. These decisions are critical; the future of many jobs
depends on making the right decisions.
We hope this topic will guide you to an enlightening and profitable future.
