Compared to cloud computing,
where we access centralized services, edge
computing refers to decentralized data processing where computing is
done at or near the source of the data, instead of depending on the cloud which
is diverts the traffic to one of the data centers to do all the processing. But
that doesn’t mean the cloud will disappear, instead the cloud is coming closer to
you.
A simple example…for
instance, if you have one Pressure sensor device which continuously sends pressure
readings to the cloud. However if you have 100 x Pressure sensor devices across
a factory/warehouse, you have a bandwidth problem. But if the sensor devices
are “smart enough” (possibility for AI / ML) to only send the important readings
and discard the rest, your internet bandwidth is saved.
One more
example would be Alexa….with local
speech recognition, Alexa only sends text to the cloud rather than voice
recordings which results in much faster response, especially for commands that
can be handled without leaving your home (e.g., turning on the light).
There is a view that there
isn’t much growth left in the cloud space as nearly everything that could have
been centralized has been already centralized, which then creates new
opportunities for bringing the cloud
closer to the edge.
The idea of edge computing is to process data on the spot if not fully at
least partially at the end device to save resources (internet bandwidth,
response times, etc.) while still benefiting from the advantages of the cloud.
At an enterprise level when we
discuss edge to cloud, they typically are thinking about storing and processing
data generated by employees or IoT devices, at the point where it is created
and consumed (at the edge). Leveraging edge to cloud architectures can result
in cost savings, improved productivity, without compromising on application
responsiveness. This requires the use of resources that are not permanently
connected to the network, such as sensors, tablets, controllers, etc.
IoT connected devices can best use the edge
computing architecture. With remote sensors installed on machines or devices,
they generate considerable amounts of data. If that data is sent across a long
network link to be analyzed, tracked & processed, that takes much more time
than if the data is processed at the edge, close to the source of the data.
With edge computing, security also
improves as encrypted files are processed closer to the network core. Though edge
computing supports real time requirements on the IoT, however for processing or
storing large amount of data it needs the power of cloud computing.
The Amazon Web Service’s Snowball
Edge device is a device with on-board storage and compute power for select AWS
capabilities. Snowball Edge can carry out local processing and edge computing
workloads and also transfer data between your local environment (devices,
sensors, tablets, etc) and the AWS Cloud. AWS Snowball Edge brings the power of
the AWS Cloud to your on premises location, which can cut down on the bandwidth
because we can either do all of the processing on the device or pre-process it
before sending it on to the cloud.
Edge to cloud file services enable
companies to manage their data centrally in the cloud, while at the same time
making that data instantly available to users at the edge. By having edge to
cloud file services, enterprises can improve application responsiveness and
availability. This enables agile collaboration, while also providing data
security, governance and compliance. Edge to cloud file services will be a key
enabler for business success by offering low latency, reliable access to data &
power of cloud-scale economics.