Industry executives and experts share their predictions for 2022. Read them in this 14th annual VMblog.com series exclusive.
Changing the Data Gravity Paradigm in 2022
By Steve Wallo - CTO, Vcinity
The past two years have certainly been challenging for many
organizations and predicting upcoming advancements in technology at this point
can only be based on very recent user experiences that, in my case, I have been
observing in our customer base. A major trend that I would like to bring up is
removing the necessity of having to collocate compute/application to their
associated data in order to get better performance and a good user experience.
The gravity problem associated with data and applications is changing for the
better. Below are some of the things that I see possible in the near future.
- Now as applications are being able
to fully move between cloud(s) and edge/premise locations, we will finally see
the data no longer having to follow the apps as they move. This has always been
the shortcoming of these gains. With the data having to follow the app, these
larger data sets consume so many resources and time it fundamentally minimizes
the advancements. I see this issue being resolved. With recently created
technology, the age-old paradigm of the compute engine and data resource needing
to be collocated no longer exists. Now dynamic and agile application movement
can be done while the data stays in place giving the apps full access and
performance without WAN penalties. This finally enables the initial vision of a
true multi-cloud and hybrid cloud environment from both the application and
data perspective. This new paradigm will allow vast improvements in business
resilience, workload efficiencies and user experience.
- With the ability for compute to
point to remote data, we can see the upcoming implementation of federated
compute power processing as a single distributed workload. Now we can leverage
compute across multiple clouds and multiple edges - all working together in
real-time to solve a single problem. In addition to working off a common
dataset, each processing node, regardless of location, can now share its own
data product to the federation of nodes. Imagine a collective of processing
instances brought up as needed pulling in from hundreds s or thousands of small
devices across a smart city. This could
empower AI workloads in the event of a natural disaster or major event for
solving/tracking problems as they unfold. This would enable immediate insights
for faster actions and decisions without having to wait for perishable copies
of data to move distances.
- As 5G is rapidly evolving, we can
see new microservices running at the near and deep edge all with coherent
access to each other's data. Want to do geo-distributed AI/ML workloads with
full awareness running across vast distances leveraging all data
simultaneously, this is something we see becoming more and more prevalent.
Imaging having any cloud accessing federated deep edge-based compute and its
associated data for true cloud-edge data fusion. The era of the super powerful but
"stranded" edge will be a thing of the past. We see this driving more advanced
edge technologies now that instant accessibility from any edge to any other
edge or cloud is becoming possible.
- One of the other things we see
changing is the role of data management. Not that data management is not
important, it is and will continue to grow as data grows, but having to manage
multiple copies of the same data scattered around the globe will subside. One
of the bigger problems and costs is maintaining, storing and managing
distributed data copies. Which one is correct, is it still needed and what
about the storage to house it? These are all common questions and concerns.
There is also the security element of multiple copies; the more copies of the
same information, the larger the attack vector for it to be compromised. With
the advent of common geographically accessible data sets the number of copies
is minimized. This will reduce one of the complexities of managing enterprise
data, as well as reduce the security target area of information.
As you can see, breaking the age-old paradigm of data
needing to be near the compute to be used is changing. The concept of data
gravity has been altered, allowing for faster, more immediate ways to collect
information from data anywhere at any time. We should all be watching for new,
interesting use cases coming out of this
and what users will be sharing with us next year.
##
ABOUT THE AUTHOR
Steve Wallo currently serves as Vcinity's CTO, overseeing
resources related to the insertion of advanced technologies and strategies into
customer architectures and future IT decision methodologies. He is responsible
for bridging future IT trends into the company's existing portfolio
capabilities and future offerings. Prior to Vcinity, Wallo was the CTO at Brocade Federal,
responsible for articulating Brocade's innovations, strategies and
architectures in the rapidly evolving federal IT space for mission success.
Wallo has served the U.S Government as
the chief architect for the NAVAIR Air Combat Test and Evaluation Facility High
Performance Computing Center.