At a Glance
- HPE pits its $14 billion acquisition of Juniper Networks as a way to revolutionize AI networking.
Juniper Networks, a major vendor of networking equipment, is being acquired by HPE in a deal worth $14 billion to create next-gen architectures to power AI.
HPE CEO Antonio Neri spoke about the deal on stage at Mobile World Congress in Barcelona. He said the combination would give the industry a “new modern architecture” driven by AI that would encompass the cloud all the way to the edge.
“Together we're going to have all the intellectual property to really converse the network, to automate the network and then ultimately, the killer application for that is AI.”
Neri was joined on stage by Juniper CEO Rami Rahim. Both he and Neri described AI as the next big inflection point, with Rahim adding that the technology is “definitely going to be bigger than the internet inflection that happened several decades ago.”
“This combination to me is about acceleration of our innovation to capture the full AI opportunity from silicon to systems to software,” Rahim said.
The acquisition would double HPE’s networking business. If it goes through, HPE could become a big challenger to Cisco, the network switches giant. Notably, Juniper began substantially investing in AI when it acquired Mist Systems in 2019. Mist offers intelligent wireless network products that use machine learning.
At MWC, the pair positioned the deal as one that will capture the opportunities of AI through modernized infrastructure. For example, HPE CEO Neri said it will disrupt business and provide them with new ways to improve productivity.
“Fortune favors the bold,” Neri said. “You have to have the courage, the conviction and belief in what you think is going happen, and then go for it. And this is what we're doing together” with Juniper.
Why buying Juniper was 'important'
In a separate interview at MWC, Neri said the acquisition would accelerate deployment of AI through the network. “Ultimately, you need an AI-native architecture that brings the data to the right computational resources to accelerate business productivity and data insights.”
Neri said most of the AI action now is on training of foundation models. But for enterprises, the reality is that they will take an existing model, fine-tune it and give it context with their own data.
In a cloud-native world, there will be thousands of workloads running on thousands of servers and everything is shared, he added. But in an AI-native world, there is one model or workload and thousands of accelerated compute.
“So the network becomes the core component of that in addition to the accelerator. That’s why this Juniper acquisition is so important,” Neri said.
Read more about:
Conference NewsAbout the Author
You May Also Like