Scaling Hyperlocal Weather Forecasts with IBM Cloud
Key Points
- The Weather Company maps the atmosphere to deliver hyper‑local, one‑kilometer‑grid forecasts on demand, serving millions of users and handling spikes from 30 million up to over 100 million during severe weather.
- Its forecast‑on‑demand system processes about 250 billion forecasts daily and supports an API platform that handles roughly 150 000 requests per second, because timely data can be a matter of life and death.
- After a six‑month migration to IBM Cloud Kubernetes, the company achieved an ~80% improvement in DevOps workflow efficiency and gained the ability to scale infrastructure instantly as weather events intensify.
- The managed IBM service also provides automated security alerts, enhanced automation, and global cloud reach, enabling developers to focus on new features and expand safety‑critical weather services worldwide.
Full Transcript
# Scaling Hyperlocal Weather Forecasts with IBM Cloud **Source:** [https://www.youtube.com/watch?v=SL-Egc0XuIU](https://www.youtube.com/watch?v=SL-Egc0XuIU) **Duration:** 00:02:47 ## Summary - The Weather Company maps the atmosphere to deliver hyper‑local, one‑kilometer‑grid forecasts on demand, serving millions of users and handling spikes from 30 million up to over 100 million during severe weather. - Its forecast‑on‑demand system processes about 250 billion forecasts daily and supports an API platform that handles roughly 150 000 requests per second, because timely data can be a matter of life and death. - After a six‑month migration to IBM Cloud Kubernetes, the company achieved an ~80% improvement in DevOps workflow efficiency and gained the ability to scale infrastructure instantly as weather events intensify. - The managed IBM service also provides automated security alerts, enhanced automation, and global cloud reach, enabling developers to focus on new features and expand safety‑critical weather services worldwide. ## Sections - [00:00:00](https://www.youtube.com/watch?v=SL-Egc0XuIU&t=0s) **Hyperlocal Forecast Scaling on IBM Cloud** - The weather firm explains its goal to provide kilometer‑resolution, on‑demand forecasts to tens of millions of users—processing 250 billion forecasts daily and 150 k API calls per second—and how a six‑month migration to IBM Kubernetes delivered 80 % DevOps efficiency and seamless rapid scaling during extreme weather events. ## Full Transcript
the weather company's mission is to map
the atmosphere of the park and based on
that generate the most accurate and
hyperlocal forecasts which can be served
out to all our consumers and devices
which number in millions across the
globe weather is unpredictable and we
need to be able to spin up and spin down
as fast as weather happens our normal
day-to-day load is 30 million unique
users and when we peak into severe
weather that can go north of 100 million
across our products so we generate
forecasts on demand and our forecasts on
demand system is responsible for
generating forecasts at a resolution of
one square kilometer grade we deliver
forecasts on demand to the tune of 250
billion per day on the backside of that
is our API platform and that system
operates at the scale about 150,000
requests per second if our products
don't work and don't work fast people's
lives are in danger
imagine a hurricane is approaching east
coast of the United States as it is
approaching the coast usage of my
weather data goes up tremendously it can
go up from anywhere from two times to
five times it took us six months to
migrate and to IBM kubernetes from our
previous provider and what we've noticed
is some real efficiencies with DevOps
it's produced our workflow in pipeline
bar about 80 percent with the latest
hurricane we were able to scale with IBM
cloud very easily very seamlessly ibm's
community service allows us the scale as
fast as the weather happens as we
migrate it our web platform to IBM cloud
kubernetes services one of the biggest
advantages is that this is a managed
service which allows our team not to
have to babysit our system enables them
to do other work with the migration to
IBM cloud kubernetes we've gained the
ability to have automation security
began this was a feature in the
component that was very male for us in
the past
IBM security team proactively now
notifies us
security vulnerabilities the the tools
and the investment that the companies
made gives us capabilities that we would
have never had before it gets us very
excited and it gives our developers and
engineers opportunities to build new
things that they wouldn't have otherwise
been able to do the global reach that
IBM and the IBM cloud together with our
technology with our capabilities for
forecasting give us the ability to
extend our products and services to keep
people safe all over the world
[Music]