Sergey Nivens - Fotolia

Kafka users detail real-time data benefits

Confluent introduces a series of new real-time data efforts as users of the open source data-streaming technology outline real-world deployment applications.

Kafka event streaming technology is helping automakers produce fast cars and motorcycles.

At the Kafka Summit Europe virtual conference on May 12, several major corporations outlined how they are using Kafka, as new options emerge to make it easier to operate and deploy the technology.

Open source Apache Kafka provides data streaming capabilities that help organizations use real-time data.

At the summit, Confluent, one of the leading contributors to the Kafka project, introduced new commercial services, including a Kafka for Kubernetes offering that enables users to run the technology on premises and in cloud-native environments.

Among the users who spoke about how they use Kafka in production environments was a representative from automaker BMW Group.

In a keynote session, Felix Böhm, product manager, IoT plant solutions and data streaming, talked about how Kafka is helping the automaker improve operations.

Böhm explained that BMW Group needed to provide data to all of its applications, from different sources and across its manufacturing process. The company also wanted to decouple data from being tethered to a specific piece of hardware or system to enable more flexibility.

BMW uses Kafka to stream data and provide insight into where data is coming from, making it possible to better understand real-time events. Better visibility into real-time data can help ensure stability, which is critically important for the automaker, Böhm said.

"Stability is really key in manufacturing, as a few seconds outage or a few minutes outage can cost you quite a lot of money when the plants are shut down and people have to go home," he said.

Porsche speeds ahead with Kafka

BMW Group isn't the only automobile manufacturer that has embraced Kafka. In a user session, Sridhar Mamella, platform manager, data streaming platforms at Porsche AG, explained how the high-end car maker is using real-time data with Kafka.

"At Porsche, it's all about speed -- and fast cars equals fast data," Mamella said. "Porsche is on route to becoming a real-time company."

Mamella said Porsche has deployed Kafka on applications such as those delivering notifications to different areas of the manufacturing process.

Playing with Kafka in real time at Sony PlayStation

Naresh Yegireddi, lead data engineer at Sony PlayStation, explained how the gaming platform vendor benefits from real-time data streaming with Kafka.

Photo of Sridhar Mamella of Porsche AG, at Kafka virtual conference
Sridhar Mamella, platform manager, data streaming platforms at Porsche AG, explained how the sports car maker is using Kafka to stream real-time data updates.

Yegireddi said Sony PlayStation uses Kafka to stream data from different sources into operational data stores where the data can be analyzed. One application in which Yegireddi's team is using Kafka is to handle price comparisons across gaming publishers and vendors. In the interactive gaming industry, consumers have many choices and many online marketplaces where they buy games, and pricing is ultra-competitive.

"There is always a possibility that if we can get to know about the insights of what our competitors price matrix looks like, that will be super useful for our pricing team to come up with strategies to be competitive all the time," Yegireddi said.

Kafka going cloud native for real-time data

During the opening keynote, Confluent co-founder and CEO Jay Kreps introduced a series of new services designed to further popularize Kafka.

Kreps is one of the creators of Kafka, which was developed while he worked at LinkedIn.

Kreps noted that managing the data associated with real-time streaming can be a challenge from a data governance perspective. To that end, Confluent now has data governance services for its Confluent Cloud platform available in early access for customers. Kreps said the service integrates data quality, data catalog and data lineage functionality.

"This really expands the use of data across a company in a way that's frictionless and correct," Kreps said. "It starts to move up the stack into getting high-quality data, and not just bytes, to all the people who need it."

Kreps also revealed the general availability of the Confluent for Kubernetes service, which is intended for private cloud and on-premises deployments. With Confluent for Kubernetes, organizations can deploy a commercially supported instance of Kafka into an existing container deployment that is running Kubernetes.

"This isn't a fully managed offering; it's something that you run yourself in a private cloud environment," Kreps said. "We can help companies do this in their on-premises environments, and this can connect up into our cloud environment and link transparently to really act as one fabric for data streams."

Next Steps

Rockset advances real-time streaming data SQL queries

Kafka streaming data gets governance from Confluent

Dig Deeper on Data management strategies

Business Analytics
Content Management