Page 10-11 - Magazine.indt

Basic HTML Version

Page 10
Page 11
Simply put, a database is a collection of data. Likewise,
databases have been around for as long as computers
have existed (even a simple hard drive is an example
of a database). However, what gives form and shape
to any database is not necessarily its “container” or
housing, but rather its organizational structure. In
other words, it is the way that operating systems,
applications, and stored data are organized that truly
provides higher functionality to any database. Cloud
computing is certainly not exempted from this either;
you could argue that a cloud’s database is the most
important element in the system, as it often contains
all other crucial components (OS, virtualization, apps,
etc.).
A cloud computing database isn’t really that different
from any other, however, that might soon change. You
see, with any database, there will be a great deal more
instances where data is read, accessed, or delivered
than those where new information is written or
uploaded. Cloud computing databases are certainly
no exception. Their “reads” still heavily outweigh their
“writes”. But given the nature of cloud computing—
what it seeks to achieve and how it is being used—
there is much more data writing occurring than ever
before. Just think of all the social media sites out there
which connect directly, and sometimes indirectly, to
any number of clouds/databases. Sure, data that’s
uploaded to these databases will be viewed many
more times than it’s written, but it should be noted
that the technology has allowed for a great deal more
uploads than ever before. The demands of cloud users
seem to be having a significant effect on the way
information is arranged in databases.
It is absolutely crucial that information available via
search queries be up-to-date at any given time (this is
of course, a no-brainer). If data is not being updated
correctly, it highlights a problem with the way it is
stored or being accessed. For example, all retain-
driven businesses that rely on Internet-based sales
absolutely must have accurate data for customers to
peruse. Failure to provide such information translates
directly into lost sales.
Additionally, data must be accessed quickly, even
across extensive databases. This is one of the big
problems facing cloud computing, actually. Just
think about all the data that hits a cloud from one or
more social media sites on a daily basis, much of it
being interconnected in numerous ways. As the size
and scale of your average cloud grows (often across
multiple server farms in several areas), it is becoming
increasingly challenging to quickly locate and deliver
specific data packets. The reason for this is once again,
organization.
If data is stored across multiple servers (as is usually
the case), queuing up a search will have to make its
way through or past a number of other machines.
The end result of this multi-system transit is degraded
performance. In other words, every time a data
partition is encountered, conventional wisdom would
state that performance is negatively affected. Some
of the major cloud computing players, like Amazon,
have taken to replicated tables as a potential solution
for this quandary, which is used in their SimpleDB
solution. Replicated tables offer an alternative to
traditional indexing and even take geographical
positioning and limitations into consideration.
Arguably, the best solution is the “Shared-disk
database”, which creates a central repository that’s
devoid of the need for partitioning. Under this model,
everything seems to function much better, especially
when cloud computing is added to the equation.
Increased elasticity is also supported.
The technology and methods of constructing and
managing cloud databases are increasing day-by-day.
While there are certainly some very serious challenges
facing cloud providers in this regard, it’s somewhat
comforting to know that there are solutions on the
table which are not only capable, but are flexible as
well. In the coming years, it will be very interesting
to see what organizations, like Google, Microsoft,
and Amazon, will come up with regarding database
requirements for Cloud Computing infrastructure.
Big Data
in the sky
Ivanka Menken
Microsoft isn’t a company that’s known for profit loss. In fact, their recent $492 million shortfall is, by all
accounts, the first actual deficit they’ve accrued in their 26 years as a company. People are beginning to
wonder if this is demonstrative of some larger systemic issue or if it’s simply an isolated incident.
Microsofttakesoff intothecloud
When MS bought aQuantive, an online advertising
company, in 2007, they probably had bigger things
in mind than what’s actually transpired. According
to official reports, it is this ad business that is chiefly
responsible for the reported losses. The aQuantive
purchase was apparently a strategic move made
by Microsoft in an effort to compete with Google
in search-related advertising. Needless to say, it
appears that things didn’t work out too well for
them.
Trying to go toe-to-toe with Google on its own
turf (search engine technology and advertising)
is a big gambit, no matter how you look at it. Still,
you have to admire Microsoft’s drive, initiative, and
shrewdness. Many corporations wouldn’t dare even approach the notion of trying to compete with Google.
Maybe it’s just another sign of the times? Nearly everywhere you look, it seems that there is economic strife
(often compounded with difficult climate-related issues tacked-on). Perhaps it was only a matter of time until
Microsoft falls victim to the incredible turmoil in the markets? At any rate, they assure us that this is an isolated
incident and not anything that’s poised to negatively affect their stock price or future plans.
One thing that is certain in the midst of all this
uncertainty, however, is that Microsoft seems to
be placing a great deal of emphasis on their cloud
computing operations. Not only does MS have
strategic positions in several key areas of the cloud
vendor market, it is also eagerly developing a number
of cloud-based technologies and pushing for a more
cloud-centric approach across the board. For instance,
Windows 8 is poised to be released soon, which is
going to utilize cloud technology in ways that no
other Windows OS has attempted before (or anyone
for that matter).
This move toward placing more focus on cloud
computing should also be a wake-up call for
businesses of all shapes and sizes. The time to get
prepared for cloud adoption and transition is now. As
you may or may not be aware of, one of the best ways
to ensure that your business is capable of utilizing,
managing, and even improving upon your current
or future cloud services is through professional
development in the IT sector.
Ivanka Menken