We find ourselves in a time when certain businesses are being asked to scale to levels they never imagined. Sometimes that increased usage comes in bursts, which means you don’t want to pay for permanent extra capacity you might not always need. Today, Confluent introduced a new scale on demand feature for its Apache Kafka cloud service that will scale up and down as needed automatically.
Confluent CEO Jay Kreps says that elasticity is arguably one of the most important features of cloud computing, and this ability to scale up and down is one of the primary factors that has attracted organizations to the cloud. By automating that capability, they giving DevOps one less major thing to worry about.
“This new functionality allows users to dynamically scale Kafka and the other key ecosystem components like KSQL and Kafka Connect. This is a key missing capability that no other service provides,” Kreps explained.
He points out that this particularly relevant right now with people working at home. Systems are being taxed more than perhaps ever before, and this automated elasticity is going to come in handy, making it more cost-effective and efficient than was previously possible.
“These capabilities let customers add capacity as they need it, or scale down to save money, all without having to pre-plan in advance, ” he said.
The new elasticity feature in Confluent is part of a series of updates to the platform, known as Project Metamorphosis, that Confluent is planning to roll out throughout this year on a regular basis.
“Through the rest of the year we’ll be doing a sequence of releases that bring the capabilities of modern cloud data systems to the Kafka ecosystem in Confluent Cloud. We’ll be announcing one major capability each month, starting with elasticity,” he said.
Kreps first announced Metamorphosis last month when the company also announced a massive $250 million funding round on a $4.5 billion valuation. In spite of the current economic situation, driven by the ongoing pandemic, Confluent plans to continue to build out the product, as today’s announcement attests.
By Ron Miller