F5 & Nvidia’s Strategy to Enhance AI Application Delivery
AI has created unprecedented demand for advanced computing technologies and infrastructure.
However, as organisations worldwide build out AI capabilities, they face significant challenges in managing the massive data flows required for training large models and delivering AI-powered applications at scale.
As a result, technology providers are developing new solutions to optimise AI workloads and enable more efficient deployment of AI systems.
In response to these emerging needs, F5, a company specialising in application security and delivery, has announced a new product aimed at addressing the infrastructure challenges posed by large-scale AI deployments.
The company has partnered with Nvidia, a leader in AI computing, to create an integrated solution that promises to enhance the performance and efficiency of AI applications.
Streamlining AI data traffic
F5's new offering, called BIG-IP Next for Kubernetes, is designed to provide a centralised control point for managing data traffic in AI infrastructures.
“Realising the potential of AI requires more data processing capabilities than the industry had previously prepared for. For many companies, deploying cutting-edge AI requires massive infrastructure buildouts that tend to be very complex and expensive, making efficient and secure operations more important than ever.”
Kubernetes is an open-source platform that automates the deployment, scaling and management of containerized applications.
The solution leverages Nvidia's BlueField-3 data processing units (DPUs) to improve the efficiency of data centre operations critical for AI workloads.
According to F5, BIG-IP Next for Kubernetes integrates networking, traffic management and security functions to help organisations maximise resource utilisation in their data centres.
This approach aims to optimise AI application performance while improving overall infrastructure efficiency.
The solution is built specifically for Kubernetes environments, which are widely used for deploying and scaling containerised applications.
F5 states that BIG-IP Next for Kubernetes has been proven in large-scale telecom cloud and 5G infrastructures and is now tailored for AI use cases such as inference and retrieval-augmented generation.
The integration with Nvidia BlueField-3 DPUs is designed to minimise hardware footprint, enable granular multi-tenancy and optimise energy consumption while delivering high-performance networking, security and traffic management.
This combination allows both mobile and fixed-line telecom service providers to ease the transition to cloud-native infrastructure, addressing the growing demand for vendors to adapt their functions to a cloud-native network functions model.
Enhancing AI infrastructure management
By offloading data-intensive tasks to Nvidia's BlueField-3 DPUs, the F5 solution aims to free up CPU resources for revenue-generating applications.
This capability could be particularly beneficial for telecom service providers transitioning to cloud-native infrastructure models.
Kunal Anand, Chief Technology and AI Officer at F5, explained that organisations are building highly optimised environments to train large AI models and deliver inference capabilities at scale.
He says: “The synergy between F5’s robust application delivery and security services and Nvidia’s full stack accelerated computing creates a powerful ecosystem.
“This integration provides customers with enhanced observability, granular control, and optimised performance for their AI workloads across the entire stack, from the hardware acceleration layer to the application interface.”
F5 claims that customers will be able to automate the discovery and security of AI training and inference endpoints, while also addressing data integrity and encryption requirements.
“WWT clients will be able to benefit from greater data ingestion performance and GPU use during model training and better user experiences during inference, while gaining a strategic control point for security services” says Todd Hathaway, Global Practice Manager, AI, App and API Security Solutions at WWT.
“Technology from F5 and NVIDIA – two of our most strategic partnerships – further strengthens our Global Cyber mission to deliver digital security excellence.”
It seems that solutions like F5's BIG-IP Next for Kubernetes with Nvidia BlueField-3 DPUs are likely to play a crucial role in enabling organisations to deploy and manage AI applications more effectively.
The focus on optimising data traffic, enhancing security and improving overall infrastructure efficiency addresses key challenges faced by enterprises and service providers in the AI era.
Kuba Stolarski, Research Vice President, Computing Systems Research Practice at IDC concludes: “Realising the potential of AI requires more data processing capabilities than the industry had previously prepared for.
“For many companies, deploying cutting-edge AI requires massive infrastructure buildouts that tend to be very complex and expensive, making efficient and secure operations more important than ever.”
******
Make sure you check out the latest edition of Technology Magazine and also sign up to our global conference series - Tech & AI LIVE 2024
******
Technology Magazine is a BizClik brand
- How Samsung's AI Ambitions Have Driven Chip FocusDigital Transformation
- How Toyota & NTT use AI to Create a Zero-Accident SocietyAI & Machine Learning
- Why CFOs are Prioritising Tech & Marketing InvestmentsDigital Transformation
- Why the Finance Sector Grapples with Software Security DebtCloud & Cybersecurity