How Samsung SmartThings Uses Sumo Logic to Monitor and Troubleshoot IoT Devices

and at smart thing is what we have to
do is we aim to be an open platforms the Internet of Things so we work with
hundreds of devices aren’t platform that we support as well as a lot of
third-party providers including nest ecobee Hugh we get with not only
zippy and z-wave protocols but Wi-Fi devices as well being a lead on the
platform reliability performance to the biggest value that gives me directly is
visibility so one of the things that we have is a bunch of different dashboards
that show us metrics all over the platform including executions errors top
smart apps or devices that people aren’t using so that we can track trends over
time to see what the most popular apps and devices are that helps us to
preemptively scale at the platform or deal with load issues that might arise
because of that and not only that we get a lot of value from other parts of the
organ cluding customer support who gets a ticket then they can search that user
and sulmona find the error and then put that stack trace into a ticket and
assign to a developer also we do a lot of stuff with product because the
product they want to know what are the most active use cases that people are
using it for like what are people coming to smart things for do they have hue
lights or Amazon echoes or what are they what’s the big draw or what’s the most
active things people are using so some of our dashboards show at the top 10
smart apps or device types or use cases that people are installing on our
platform which is heavily used by product as you can imagine to track
maybe four top apps we want to add more functionality or if things start to
trend downward we want to know why and we can fix issues or problems that maybe
you’re making knows they’re not as popular as they once were we use
simulate across a variety parts of our organization including customer support
developers themselves we have a lot of product dashboards as well for a
trending smart apps or device types as well as even we call them I guess I call
them exec level dashboards so maybe it’s a bunch of metrics that we want to
add a big similar parse for and boil that down to like a pass/fail or good
great failing tip of metric so will to show a panel that would expose all those
metrics boil it down to simple terms that anyone can understand and people
can follow those dashboards daily or trending over time to keep a pulse on
the stability and performance of the platform one of the biggest valuates at SEMA has
had that no other product and the space has had for us has been the adoption
rate so we have over 260 employees and we have over 250 symbol counts that
means we have 99 95 plus percent of our users are actually actively using Simo
so like I said across all of our orgs all of our functions we have heavy heavy
super users as well as just daily active users that are doing searches constantly
looking for different data that they can mine and pull out whatever value for
their context or team that makes sense for them so adoption has been huge for
our customers and our community is really the heart of smart things that’s
what the platform is all about is how do we identify problems quickly so that we
can resolve them so that there’s mitigated impact to the platform
and their use of our system so if we can identify those as early as possible or
mediate them before there is ever an issue sometimes even before they notice
there’s a problem that’s ultimately the goal so we have a lot of dashboards that
track latency over time exceptions over time even particularly users if they’re
stuck trending in errors so that we can reach out to them or fix them quickly

Bernard Jenkins

Leave a Reply

Your email address will not be published. Required fields are marked *