18 / 58
edfas.org
ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 18 NO. 3
18
MANAGEABILITY CHALLENGES FOR
INTERNET OF THINGS
Yen-Kuang Chen, Principal Engineer, Intel Corporation, and Associate Director,
Intel-NTU Connected Context Computing Center
y.k.chen@ieee.orgEDFAAO (2016) 3:18-21
1537-0755/$19.00 ©ASM International
®
T
here is no longer much argument around the
Internet of Things (IoT) concept as the “next big
thing,” but consensus remains elusive around the
next level of questions anddiscussion: Why is the IoT going
to be so great, and what are the obstacles to achieving
that vision?
The IoT is already delivering valuable benefits in the
nascent stage of its development. However, I would argue
that “IoT Version 1.0” has not yet been realized, and
the magnitude of this revolutionary innovation will not
become clear until then.
Getting to that point will require addressing a variety
of user pain points, perhaps the most glaring of which is
device failure. Mainstream adoption of and reliance on
the IoT demands a scenario in which one or more dispa-
rate devices may fail but the overall system continues to
function. Perhaps the system would not function as well
as when the failed devices were operational, but it would
continue. In the meantime, the failed device could be
recognized and repaired without the user experiencing a
disruption in service and with a return to optimal system
performance.
In addition to more functionally reliable devices,
IoT 1.0 will require an intelligent middleware layer for
multivendor device management. Achieving such a layer
will require global collaboration across the IoT’s diverse
stakeholders.
WHEN WILL THE IoT REALLY BE
THE IoT?
I began working on the IoT six years ago, and at that
time, I didn’t have a clear definition of what the IoT is or
would be.
Even today, different people have different definitions
of the IoT. For some, the IoT is having things connected to
a smartphone and enabling capabilities such as remotely
locking/unlocking the front door. For others, the IoT is
having a device connected to the internet, streaming data
to the cloud, andhaving the cloudperform intelligent ana-
lytics tohelphumansmake intelligent decisions. However,
such definitions are still predicated on the notion of a
small number of devices connected through the internet
to individual humans, who, at the very least, are kept in
the loop for all the real decision-making.
My definition of the true IoT—IoT 1.0, if you will—is
when heterogeneous, multiple connected devices are
working together to our benefit and without us having to
make all of the decisions (Fig. 1). The IoTwill deliver more
benefits with more and more devices working together
without human interaction, whichbothnaturally impedes
IoT scalability and adds complexity to our lives.
For example, mental wellness is an area that especially
interests me. In the IoT 1.0, devices that measure various
bio signals, such as heart rate and temperature, could be
Fig. 1
When heterogeneous, multiple connected devices are working together without humans making all of the decisions,
we will attain IoT 1.0.