HPE and Oracle expand networking partnership for gigawatt-scale AI data center buildout

HPE and Oracle Announce Multi-Year AI Infrastructure Partnership

HPE and Oracle have expanded their networking collaboration to support one of the industry's largest AI infrastructure buildouts, with HPE Juniper Networking equipment set to be deployed across Oracle's AI data centers globally1

1

2

2

. HPE CEO Antonio Neri announced the deal during HPE's third quarter earnings call, describing it as a multi-gigawatt networking infrastructure deployment that positions HPE uniquely in the AI data center market2

2

.

The expanded agreement builds on more than a decade of engineering work between Oracle and Juniper Networks, which HPE acquired last year4

4

. As part of the agreement, HPE issued Oracle warrants to purchase shares of HPE common stock1

1

.

Deployment Across Oracle Cloud Infrastructure's AI Superclusters

The multi-year networking deal covers the deployment of HPE Juniper Networking PTX and MX routing platforms and QFX and EX switching platforms across Oracle Cloud Infrastructure's data centers, along with multi-year networking support services and financing capabilities1

1

3

3

. HPE Juniper Networking equipment already supports key elements of Oracle Cloud Infrastructure's data center and edge networks, providing the scale, performance, and availability needed for Oracle's global cloud operations3

3

.

The expanded collaboration specifically targets Oracle's growing AI supercluster buildout, where large GPU clusters place heavy demands on bandwidth, latency, congestion management, and fault recovery1

1

. Mahesh Thiagarajan, executive vice president of Oracle Cloud Infrastructure, stated that HPE Juniper Networking now plays a key role in Oracle's AI clusters, regional data centers, and edge networks with the scale, performance, and reliability customers expect3

3

.

Advanced Technology for AI Workloads

The latest QFX switches will provide high-density connectivity, dynamic load balancing, and advanced congestion management for large AI backend networks using RoCEv2, or RDMA over Converged Ethernet version 21

1

4

4

. Antonio Neri emphasized that the gigawatt-scale AI infrastructure will use HPE's QFX data center switching based on Broadcom's Tomahawk silicon as part of a scale-out architecture2

2

.

The PTX routing platforms are based on HPE's proprietary Express 5 ASIC, which Neri described as a major differentiation for scalability. He noted that only two or three players have achieved this capability at massive scale, making it a unique source of competitive advantage2

2

. HPE was notably the first to market with 1.6 terabit routing capabilities featuring both air-cooled technology and direct liquid cooling2

2

.

Intelligent Telemetry for Enhanced Network Visibility

To help Oracle Cloud Infrastructure sustain high performance and resiliency as its AI infrastructure scales, Oracle and HPE are collaborating on intelligent telemetry that delivers deeper network visibility into device and fabric behavior3

3

. Earlier detection of queue buildup, packet loss, traffic imbalances, and component degradation can help Oracle resolve issues faster and minimize their impact on GPU utilization and network availability4

4

.

Praveen Jain, senior vice president and general manager of data center and AI networking at HPE, emphasized that every microsecond matters to AI models and agents, and the high-performance networking combined with deeper fabric visibility helps Oracle detect bottlenecks faster and improve the efficiency of AI infrastructure3

3

.

Market Implications and Future Outlook

Antonio Neri expects the Oracle deal to act as a high-profile network infrastructure buildout that could propel more networking AI deals for HPE. He stated that this proves HPE has a scalable set of products that deliver the performance with AI capabilities and self-driving network operations that the market is seeking2

2

. Neri highlighted that HPE has embedded AI capabilities across the entire portfolio, not just in campus and branch networks, with routers capable of optimizing bandwidth using AI2

2

.

Rami Rahim, executive vice president and president of Networking at HPE, noted that Oracle is building AI infrastructure at extraordinary scale, and HPE is uniquely positioned to support Oracle Cloud Infrastructure with one of the industry's most comprehensive networking portfolios for AI data centers3

3

. HPE is betting the Oracle deal drives significant interest in its networking infrastructure portfolio as organizations look to scale their AI workloads2

2

.