Hewlett Packard Enterprise (HPE) has made a sweeping set of announcements at its Discover conference in Las Vegas, unveiling a broad range of new hardware, software, and integration initiatives designed to help enterprise customers build and manage large-scale AI infrastructures—from the data center to the edge. The moves signal the company’s determination to leverage its acquisition of Juniper Networks and to position its portfolio as a comprehensive foundation for agentic AI and autonomous networking.
New QFX switches target AI inferencing and scale-up architectures
At the center of the hardware launch is the HPE Juniper Networking QFX5140 switch, a 1RU fixed-configuration data center switch that delivers 16 Tbps of throughput. It is purpose-built for AI inferencing and edge AI use cases, filling a mid-tier gap in the QFX family between the high-end 102T QFX5240/QFX5250 models and the entry-level 100GbE QFX5100 series. The QFX5140 supports 24x 400G QSFP112 ports, 8x 800G OSFP800 ports, and 2x SFP28 ports, and includes support for RDMA over Converged Ethernet (RoCEv2). Its congestion management features—such as Priority Flow Control, Explicit Congestion Notification, and dynamic load balancing—are critical for efficient GPU-to-GPU communications in AI clusters, according to HPE CTO Fidelma Russo.
HPE also introduced the QFX5252 module, designed for its 72-GPU-per-rack AMD Helios turnkey AI platform. This module combines CPUs, GPUs, and open Ethernet networking into a unified high-end system aimed at both training and high-volume inferencing workloads. The QFX family is now integrated into HPE’s Data Center Director management platform, giving customers a centralized view of all network components to improve visibility and speed troubleshooting.
Deepening Mist AI integration across the portfolio
Continuing the integration theme, HPE announced that its Mist AI engine—originally associated with Juniper’s wired and wireless access products—will be extended to the HPE Aruba Central management platform and vice versa. The core AIOps engine, Marvis, now collects telemetry and user state data from Juniper routers, switches, access points, firewalls, and applications to detect and resolve a wide range of enterprise networking problems. A key component, Marvis Actions, uses AI to identify and prioritize network problem remediation. This capability will be extended to Aruba Central by the end of the year.
In addition, HPE is integrating the HPE Aruba CX switching portfolio with Mist, giving CX customers AI-native visibility, zero-touch provisioning, wired assurance for Layer 2 access, service-level insights, and HPE Marvis AI-driven support. The company also expanded Mist’s data center capabilities to include predictive analytics for proactive maintenance—for example, predicting potential optics failures before they cause outages.
Perhaps the most notable advancement is the use of an advanced reasoning AI agent within Mist for high-confidence remediation. This agent continuously and autonomously reasons across diverse data streams, including millions of Technical Assistance Center (TAC) cases and a contextual graph database from HPE Networking Data Center Director, to deliver precise root cause analysis inside the data center. HPE’s Rami Rahim described this as “the self-driving network moving from the campus to the data center,” drastically reducing the time needed to diagnose and resolve issues.
Industry analysts see this as a strong validation of HPE’s acquisition strategy. Mike Leibovitz of Gartner noted that “agentic NetOps is the most exciting area of innovation in enterprise networking in more than 20 years,” and that HPE is well-aligned by expanding Marvis across its portfolio. He added that many enterprises will start this journey with existing vendors like HPE, Cisco, or Arista, but the market is moving quickly and leadership is still up for grabs.
Unified SASE platform simplifies WAN and security management
HPE also announced a new SASE Orchestrator package, which ties together its SD-WAN and SSE offerings with cloud security and a unified policy engine. The orchestrator uses AI to manage branch, remote user, and cloud connectivity from a single interface. Customers can set security policies once and deploy them across many sites, simplifying zero-trust adoption and improving user experience through intelligent traffic steering and application awareness.
Nvidia partnership enhancements for AI agent management
HPE has deepened its integration with Nvidia, particularly through the HPE Private Cloud AI platform—a turnkey AI factory co-engineered with Nvidia. The platform now includes support for Nvidia’s Agent Toolkit software, including Nemotron open models, NemoClaw, and OpenShell secure runtime. This provides an agent operating system that reasons, monitors agent behavior, enforces policies, and reduces deployment risk. Additionally, HPE is bringing Nvidia Confidential Computing to the HPE AI Factory, protecting models and private data during execution for on-premises or sovereign deployments.
Zerto and Morpheus updates address agent security and cost mitigation
HPE Zerto Software has been updated to allow customers to identify rogue agent actions and use data protection to rewind to a clean state. The Private Cloud AI package now supports secure local agent registration, giving customers the ability to approve AI models, skills, and tools while adhering to centralized governance and security policies.
For HPE Morpheus—a platform that manages virtual machines, containers, and cloud resources across multiple environments—HPE announced that customers who buy or already own HPE’s VM Essentials package for a year will receive the first year of licenses for free. This includes Zerto migration licenses to help move workloads from legacy platforms, addressing the “double-bubble cost” problem often faced during migration. Zero-interest financing for HPE cloud ops software over three years further supports customer transitions. Russo emphasized that this cost mitigation is intended to accelerate customer adoption and ease the shift from platforms like VMware.
The breadth of these announcements underscores HPE’s commitment to becoming a one-stop provider for AI-era networking, from the data center core to the edge and the WAN. By weaving together Juniper’s networking technology, Mist’s AI engine, and a growing ecosystem of partners like Nvidia, HPE is positioning itself as a key player in the rapidly evolving landscape of autonomous, AI-driven enterprise networks.
Source:Network World News