EDGE COMPUTING & AI
Storing data – why organisations
need to rethink their approach
ABy Peter Ruffley ccording to research group IDC, the number of connected
devices is forecasted to grow to 42bn by 2025. With
the demand for the Internet of Things (IoT), automation
and 5G continuing to grow, and heavily influencing businesses
and supply chains over the coming years, the sheer volume of
data that companies will be dealing with will become more and
more overwhelming. Whereas five to ten years ago we’d see
new data centres popping up everywhere to store and move
all of the data around, this is no longer the case. Many cities,
such as Amsterdam, have put a stop to anymore data centres
being built as they drain power from the grid and cities have to
invest more in power and cooling systems to keep them running
efficiently.
Keeping the lights on
Many companies continue to operate on the ethos that nearly
all of their data, if not all of it, should be kept as there might be
some value in it, either now or in the future. However, whilst in
many aspects the cost of storing data has dropped substantially,
making it a lot cheaper to store more, it’s not declining at the
same rate as the increase in the amount of data being created.
The type of data that is actually proving to be mission-critical is
also a lot less than organisations first thought. So, can businesses
start to look at ways to not aggregate the data, but only
keep that which is necessary and valuable to their business?
How do they ensure they hold data for the right period of time
for regulatory reasons, but no longer, and how can they make
sure they have all the data they desperately need? Edge computing
is undoubtedly going to have a substantial role to play
in helping organisations become more efficient and help them
better utilise their resources. Analysing data at the point where
it is created and limiting the amount of data being sent on to the
cloud will also ease pressure on the cloud network.
An overarching change is required
The Cloud has to date, been heralded as the saviour of data
storage, with businesses only using the resources they need at
any one time. Any data that isn’t needed is put into low cost,
‘third tier’ areas of storage. However, the Cloud has proven to
not be an efficient way of allowing people to store, manipulate
and manage their data, and is in fact turning out to be an expensive
option for many, with it costing them to put data in and
store it as well as take it out and move it.
There is a misconception that having more servers (effectively
‘throwing more tin at the problem’) is the way forward.
While they are able to store large volumes of data, they do not
reduce the power needed and increase cooling costs in the
data centre, and only a few of the capabilities of these servers
are ever fully utilised. Why are businesses buying more capability
than they really need or those that they will never use? Either
way, this isn’t really getting the crux of the issue. These and
other smaller changes just aren’t enough and will not make the
intended impact. There must be smarter initiatives put in place.
Gartner predicts that by 2025, 75% of enterprise generated
data will be created and processed outside a traditional centralised
data centre or cloud. Whilst we are going to need a lot of
them to cope with the growing influx of data, edge devices are
small and they’re more compact and efficient. But how do we
do that?
Regulatory and environmental concerns
Many businesses sight regulatory requirements as a reason to
collect and keep all of their data but this is simply not justifiable.
There is a limited amount that industries and businesses truly
need to keep for these reasons.
There’s no getting away from the fact that more data means
more infrastructure to help store and move it from one location
to another. Moving data can not only be costly but due to the
amount of energy used in transferring it but there is also the
environmental impact that must be considered.
The best time for businesses to decide what they are going
to do with their data is when it is first collected on an edge /
connected device - when it’s as current, complete and meaningful
as it is ever going to be. Companies need to make a decision
at that point about what they’re going to do with it - whether
they’re going to keep it, change it or add to it. Just keeping it
has no use. Keeping a lot of data drives a lot of waste.
5G is a very short range network that, whilst it will help
transfer data faster, will need to be supported by a lot of redistribution
points (nodes) so will not help businesses until a full
Peter Ruffley is CEO at Zizo - www.zizo.co.uk
www.eenewseurope.com eeNews Europe April 2020 News 37
/
/www.zizo.co.uk
/www.eenewseurope.com
