hi and Welcome to our technical deep
dive on XM Pro stream hosts and
collections in this video we'll explore
how these powerful tools enable scalable
real-time data processing whether you're
managing a few assets or thousands
stream hosts and collections provide the
flexibility efficiency and scalability
needed to handle complex data
orchestration tasks let's get started
and see how XM Pro can transform your
DOA
operations when building intelligent
business operations Solutions it's easy
to focus solely on the user interface
however this overlooks The crucial
components that power and drive the
solution behind the scenes data streams
stream hosts and collections are the
hidden yet essential elements of XM Pro
ensuring your intelligent operation
Solutions are robust and
scalable the data stream designer is a
canvas where we map out real-time data
flows for specific use cases it allows
users to integrate transform orchestrate
and sanitize data within a single
cohesive data workflow with over 100
Integrations 75 plus analytics agents
and 60 plus action agents it empowers
you to derive meaningful insights
predict future Trends and automate
responses
seamlessly underpinning our data streams
are stream hosts which are responsible
for executing the logic of the data
streams configured in the data stream
designer when you configure a data
stream and hit publish a stream host
takes over to run it the number and
location of stream host depend on your
us case for centralized operations you
might use a single stream host for
localized operations stream hosts can be
deploy as close to assets as required
ensuring real-time data processing right
where it's needed stream hosts run the
data streams designed in the data string
designer executing the logic and
processes you have configured they
enable the agent library to communicate
with various entities including assets
line of business applications and other
systems let's jump into an example to
consider this scalability of string
hosts imagine you're monitoring the
temperature of an asset the data stream
collects temperature data filters for
values above a certain threshold and
then creates a work order in for example
sap at the bottom we have one asset with
compute capability next to it and we
have a Stream post deployed on it this
Computing capability could be a Windows
server or a Linux based system there are
multiple options available to us here
currently this stream host is running
for only one asset but what happens when
we have multiple
assets this is where collections come in
a collection is a container that enables
a grouping of stream posts running the
same use cases a collection allows us to
deploy one data stream to multiple
stream posts running the same use case
automatically by doing this a single
data stream can be published to one
collection and the stream posts by self
subscribing to the collection download
and then run the data stream logic
next let's touch on collection variables
collection variables are common to all
stream hosts that subscribe to a
collection each stream host defines a
value for these variables
locally this setup enables the same data
stream to be deployed to different
stream posts but the connection details
it requires can be unique per Stream
post for example if you have a data
stream that requires specific
configuration settings such as IP
addresses port numbers or authentication
credentials collection variables allow
each Stream post to use its own local
values for these
settings this means that while the logic
of the data stream remains consistent
across all stream posts the specific
interactions and connections can be
tailored to the unique requirements of
each Stream
post now that we have discussed the
various aspects of stream hosts and
collections let's conclude with a few
examples of how this is deployed in
practice in this example you have
multiple assets communicating with a
single stream host which then drives the
actions various assets can send data to
a central OBC server which can be
intercepted and managed by a single
Stream post this setup works well for
centralized data management and action
execution alternatively you can create a
different topology where each plant has
its own stream host these individual
stream hosts can send data to a central
cloud-based stream host which further
processes and integrates the
information this decentralized approach
allows for low localized data processing
and better management of specific plant
operations you can also have a
configuration where each stream host
runs different data streams such as LPC
mqt or others and sends the processed
data to a Central
Area this setup ensures that each data
stream is managed efficiently according
to its specific protocol and
requirements finally there's an option
to run one Stream post per asset
assuming you have sufficient compute
capability next to each asset asset when
it comes to scalability these different
topologies whether running stream posts
next to assets using a historian or
leveraging cloud-based servers allow you
to optimize data processing according to
your operational needs you can even have
distributed streams or parts of a data
Stream Run on different stream hosts
though this requires careful
consideration of latency and other
factors now let's conclude by addressing
some frequently asked questions if you
have any more questions please reach out
to the accent pro team
and one of our solution Engineers would
be happy to assist you
