F5 expands performance, multi-tenancy, and security capabilities for fast-evolving AI landscape with NVIDIA
F5, the global leader in delivering and securing every app and API, today announced new capabilities for F5 BIG-IP Next for Kubernetes accelerated with NVIDIA BlueField-3 DPUs and the NVIDIA DOCA software framework, underscored by customer Sesterce's validation deployment. Sesterce is a leading European operator specializing in next-generation infrastructures and sovereign AI, designed to meet the needs of accelerated computing and artificial intelligence.Extending the F5 Application Delivery and Security Platform, BIG-IP Next for Kubernetes running natively on NVIDIA BlueField-3 DPUs delivers high-performance traffic management and security for large-scale AI infrastructure, unlocking greater efficiency, control, and performance for AI applications. In tandem with the compelling performance advantages announced along with general availability earlier this year, Sesterce has successfully completed validation of the F5 and NVIDIA solution across a number of key capabilities, including the following areas:
- Enhanced performance, multi-tenancy, and security to meet cloud-grade expectations, initially showing a 20% improvement in GPU utilization.
- Integration with
NVIDIA Dynamo
and KV Cache Manager to reduce latency for the reasoning of large language model (
LLM
) inference systems and optimization of GPUs and memory resources.
- Smart LLM routing on
BlueField
DPUs, running effectively with
NVIDIA NIM
microservices for workloads requiring multiple models, providing customers the best of all available models.
- Scaling and securing Model Context Protocol (
MCP
) including reverse proxy capabilities and protections for more scalable and secure LLMs, enabling customers to swiftly and safely utilize the power of MCP servers.
- Powerful data programmability with robust
F5 iRules
capabilities, allowing rapid customization to support AI applications and evolving security requirements.
'Integration between F5 and NVIDIA was enticing even before we conducted any tests,' said Youssef El Manssouri, CEO and Co-Founder at Sesterce. 'Our results underline the benefits of F5's dynamic load balancing with high-volume Kubernetes ingress and egress in AI environments. This approach empowers us to more efficiently distribute traffic and optimize the use of our GPUs while allowing us to bring additional and unique value to our customers. We are pleased to see F5's support for a growing number of NVIDIA use cases, including enhanced multi-tenancy, and we look forward to additional innovation between the companies in supporting next-generation AI infrastructure.'
Highlights of new solution capabilities include:
LLM Routing and Dynamic Load Balancing with BIG-IP Next for Kubernetes
With this collaborative solution, simple AI-related tasks can be routed to less expensive, lightweight LLMs in supporting generative AI while reserving advanced models for complex queries. This level of customizable intelligence also enables routing functions to leverage domain-specific LLMs, improving output quality and significantly enhancing customer experiences. F5's advanced traffic management ensures queries are sent to the most suitable LLM, lowering latency and improving time to first token.
'Enterprises are increasingly deploying multiple LLMs to power advanced AI experiences—but routing and classifying LLM traffic can be compute-heavy, degrading performance and user experience,' said Kunal Anand, Chief Innovation Officer at F5. 'By programming routing logic directly on NVIDIA BlueField-3 DPUs, F5 BIG-IP Next for Kubernetes is the most efficient approach for delivering and securing LLM traffic. This is just the beginning. Our platform unlocks new possibilities for AI infrastructure, and we're excited to deepen co-innovation with NVIDIA as enterprise AI continues to scale.'
Optimizing GPUs for Distributed AI Inference at Scale with NVIDIA Dynamo and KV Cache Integration
Earlier this year,
NVIDIA Dynamo was introduced
, providing a supplementary framework for deploying generative AI and reasoning models in large-scale distributed environments. NVIDIA Dynamo streamlines the complexity of running AI inference in distributed environments by orchestrating tasks like scheduling, routing, and memory management to ensure seamless operation under dynamic workloads. Offloading specific operations from CPUs to BlueField DPUs is one of the core benefits of the combined F5 and NVIDIA solution. With F5, the Dynamo KV Cache Manager feature can intelligently route requests based on capacity, using Key-Value (KV) caching to accelerate generative AI use cases by speeding up processes based on retaining information from previous operations (rather than requiring resource-intensive recomputation). From an infrastructure perspective, organizations storing and reusing KV cache data can do so at a fraction of the cost of using GPU memory for this purpose.
'BIG-IP Next for Kubernetes accelerated with NVIDIA BlueField-3 DPUs gives enterprises and service providers a single point of control for efficiently routing traffic to AI factories to optimize GPU efficiency and to accelerate AI traffic for data ingestion, model training, inference, RAG, and agentic AI,' said Ash Bhalgat, Senior Director of AI Networking and Security Solutions, Ecosystem and Marketing at NVIDIA. 'In addition, F5's support for multi-tenancy and enhanced programmability with iRules continue to provide a platform that is well-suited for continued integration and feature additions such as support for NVIDIA Dynamo Distributed KV Cache Manager.'
Improved Protection for MCP Servers with F5 and NVIDIA
Model Context Protocol (MCP) is an open protocol developed by Anthropic that standardizes how applications provide context to LLMs. Deploying the combined F5 and NVIDIA solution in front of MCP servers allows F5 technology to serve as a reverse proxy, bolstering security capabilities for MCP solutions and the LLMs they support. In addition, the full data programmability enabled by F5 iRules promotes rapid adaptation and resilience for fast-evolving AI protocol requirements, as well as additional protection against emerging cybersecurity risks.
'Organizations implementing agentic AI are increasingly relying on MCP deployments to improve the security and performance of LLMs,' said Greg Schoeny, SVP, Global Service Provider at World Wide Technology. 'By bringing advanced traffic management and security to extensive Kubernetes environments, F5 and NVIDIA are delivering integrated AI feature sets—along with programmability and automation capabilities—that we aren't seeing elsewhere in the industry right now.'
F5 BIG-IP Next for Kubernetes deployed on NVIDIA BlueField-3 DPUs is generally available now. For additional technology details and deployment benefits, go to
www.f5.com
and further details can also be found in a
companion blog from F5
.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
&w=3840&q=100)
First Post
2 hours ago
- First Post
G7 backs US proposal to exempt American firms from parts of global tax agreement
The United States and the Group of Seven nations have agreed to support a proposal that would exempt US companies from some components of an existing global agreement read more The United States and other Group of Seven (G7) nations have agreed to support a proposal that would exempt American companies from certain components of a global tax agreement, the G7 said in a statement on Saturday. The agreement follows a US proposal aimed at shielding American firms from new international tax charges on the grounds that they already pay equivalent taxes under domestic law. The G7 stated that the arrangement acknowledges existing US minimum tax regulations and seeks to enhance stability in the international tax framework. US Treasury Secretary Scott Bessent signalled Thursday that a deal is forthcoming among G7 nations allowing US firms to be excluded from certain taxes imposed by other countries. STORY CONTINUES BELOW THIS AD 'After months of productive dialogue with other countries on the OECD Global Tax Deal, we will announce a joint understanding among G7 countries that defends American interests,' he said in a series of social media posts. US President Donald Trump has pushed back on the global tax agreement, with Bessent on Thursday pointing to advances on that front. 'Based on this progress and understanding, I have asked the Senate and House to remove the Section 899 protective measure from consideration in the One, Big, Beautiful Bill,' Bessent added, referring to a bill currently before US lawmakers that would slash social program spending for tax cuts. Section 899 has been dubbed a 'revenge tax,' allowing the government to impose levies on firms with foreign owners and on investors from countries deemed to impose unfair taxes on US businesses. The clause sparked concern that it would inhibit foreign companies from investing in the United States. With inputs from agencies
Time of India
8 hours ago
- Time of India
RTL buys Sky Deutschland in deal to take on US streaming giants
European broadcaster RTL Group said on Friday it would buy Sky Deutschland , in a deal that combines Sky 's sports and streaming offerings with RTL's news and entertainment brands to create a business with 11.5 million paying subscribers. Shares in RTL surged 12% following announcement of the deal, rising to the top of Germany's mid-cap index. The merger brings together two of Europe's strongest media offerings in sport and entertainment in a battle to catch up with U.S. heavyweights in Germany. RTL and Sky's combined streaming services boast an audience larger than that of Disney but far behind market leaders Netflix and Amazon Prime. The deal includes a 150 million euro ($176 million) upfront payment plus a variable component of up to 377 million euros depending on RTL's share price which Sky parent Comcast can trigger any time within five years. The purchase gives RTL, which is majority-owned by German media group Bertelsmann , local access to Sky's premium sports rights including Bundesliga and Premier League soccer, and Formula 1 motor racing, as well as Sky's WOW streaming service. A person familiar with the situation said RTL had approached Sky for the deal, and that Sky was not looking to sell any other part of its business. RTL Chief Executive Thomas Rabe described the deal as "transformational" for the group and said it would create cost savings of around 250 million euros per year within three years of the deal closing. In the past, Rabe had considered a bid for German competitor ProSiebenSat.1 but doubted whether competition regulators would give that the green light. He told Reuters in a separate interview that RTL would not be involved in any further consolidation in the German market. "The issue of a merger with ProSiebenSat.1 is now definitely off the table," he said. ProSieben, for its part, is trying to fend off a takeover from MFE, the TV broadcaster controlled by Italy's Berlusconi family, which wants to build on its commercial TV operations in Italy and Spain to create a pan-European broadcaster. Sky Deutschland, which operates in Germany, Austria, Switzerland, was on track to break even on an earnings before interest, taxes, depreciation and amortization (EBITDA) basis, Sky Group Chief Executive Dana Strong said. The business reported around 2 billion euros in annual revenue. The variable component of the deal with Sky depends on RTL's share price, which was 35.40 euros at 0721 GMT in Frankfurt trade following a jump in value after announcement of the deal. Comcast can trigger it at any time within five years of the deal being finalised provided the share price exceeds 41 euros. At a share price of 70 euros, the additional payment for Comcast would be capped at 377 million euros. RTL can pay in cash, shares or a combination of both.
Mint
9 hours ago
- Mint
Gold price outlook: MCX gold rate may hit ₹94,000 level as Israel-Iran ceasefire, trade talks dent safe-haven demand
Gold prices on Multi Commodity Exchange (MCX) fell sharply on Friday, tracking losses in the global bullion market, as signs of easing geopolitical tensions in the Middle East with the Israel-Iran ceasefire holding up, and global trade optimism dampened safe-haven demand for the precious metal. MCX gold rate for August futures ended lower by ₹ 1,563, or 1.61%, at ₹ 95,524 per 10 grams. During the session, gold prices hit a low of 94,951, falling by ₹ 2,136, or 2.2%, from its previous close of ₹ 97,087. For the week, MCX gold price slumped 3.61%. MCX silver prices for September futures also declined. The white metal ended down by ₹ 1,468, or 1.36%, at ₹ 1,06,429 per kg. It touched a low of ₹ 1,05,380 during the session. In the international market, comex gold fell over 1.5% to trade below $3,270 per ounce, marking a second straight weekly loss and a nearly 3% decline for the week. 'The slide in gold prices comes as geopolitical risks ease and global trade optimism strengthens, dampening safe-haven appetite. A tentative ceasefire between Israel and Iran held firm, while US Commerce Secretary Howard Lutnick indicated trade frameworks with China and others are nearing completion — fueling risk-on sentiment in markets,' said Jigar Trivedi, Senior Research Analyst at Reliance Securities. Meanwhile, the US dollar dropped over 1.5% to below 97.3, its lowest since February 2022, weighed down by rising expectations of interest rate cuts following US Federal Reserve Chair Jerome Powell's dovish congressional testimony. Gold price next week will be influenced by a slew of factors, including economic data release in the US and European region. 'Key focal points for the coming week include, the progress in US trade talks ahead of the July 9 tariff decision deadline, the ECB Central Bank Forum, with comments expected from top officials including Chair Powell, and US macro data such as Nonfarm Payrolls, ISM Manufacturing & Services PMIs, and trade figures,' Trivedi noted. Gold price is expected to remain under pressure with a bearish bias, according to Jigar Trivedi. 'Comex gold has key support at $3,230 – $3,200 per ounce, while resistance is placed at $3,350 level. MCX gold price for August futures is expected to decline towards ₹ 94,800 – ₹ 94,000 per 10 grams as momentum remains weak,' Trivedi said. Read all Commodity Market news here
