HomeAway, rental website with personalized offers
High-performance computing is the ability to process data and perform complex calculations at high speeds.
You keep all your data on the cluster node and execute your kilobyte-size custom code over petabytes of data, avoiding network-like utilization.
Client applications usually bring data from servers, use the records for local calculations, and discard the data as soon as the business task is completed.
This approach does not scale well if a significant volume of data gets transferred over the network.
Apache Ignite supports a co-located processing technique. The primary aim of this technique is to increase the performance of your data-intensive or compute-intensive calculations by running them straight on Ignite cluster nodes.
In co-located processing, calculations are done on local data sets of the cluster nodes. This avoids records shuffling over the network and eliminates the impact of network latency on the performance of your applications.
— Broadcast your tasks to use all the CPUs of your distributed cluster.
— Or execute your computations on a specific group of nodes.
If some of the nodes are over-utilized, Ignite can load balance your calculations to the other node.
Some computations might take minutes or hours to complete, e.g. drug discovery or logistics planning. You don’t need to start from the very beginning if something goes wrong.
Whenever you visit Amazon, Walmart, Booking, or other websites, you see personalized content, such as relevant deals made especially for you.
Personalized content that takes your age, location, preferences, and previous interactions into consideration is processed and displayed in just a few seconds thanks to high-performance computing tasks that process gigabytes of data within a second.
HomeAway, rental website with personalized offers
To create a new vaccine or drug you have to run thousands or millions
of simulations to come up with the best formula.
E-Therapeutics uses Apache Ignite capabilities for drug discovery.
nference.ai: high-performance compute APIs used to define different statistical analyses and execute numerical data in real time.
Dutch Railways have thousands of different trains to deliver cargo and people across the country. They need to calculate how all those trains should be moving in real time. These calculations also happen with high compute APIs.
High-performance computing allows the processing of unlimited data sets,
with analysis taking only seconds.
JPMorgan Chase use Apache Ignite for heavy computations which help to make effective exposure management.
Discover our quick start guide and build
your first application in 5-10 minutes
Check out success stories from
different industries across the world