Arista Networks has taken the wraps off its 7060XE7 Series, a new portfolio of 1.6T networking platforms designed to provide the foundation for rack-scale AI infrastructure. The 7060XE7 family features fixed switch platforms and configurable rack-scale systems, targeting racks for vertical and horizontal AI workflows. All will run Arista’s Extensible Operating System (EOS), which includes low-latency and intelligent packet buffering to manage the intense microbursts typical of AI communication and collective patterns, Arista stated. The 7060XE7 family is built on Broadcom Tomahawk 6 silicon. Arista is also working with AMD on next-generation compute silicon and NICs to enable scale-out AI fabrics, the company said.

Strategically, the 7060XE7 Series signifies Arista’s transition from offering standalone, high-performance switches to providing rack-scale systems that can handle the extreme density, power, and thermal efficiency AI requires. The platforms allow customers to build scale-up and scale-out AI fabrics using air, liquid and hybrid-cooled technology. Specific configurations include:

• 7060XE7-64PS and 7060XE7-64PRS 4U Rack Switches: Available in Q4, these air-cooled systems offer support for pluggable Integrated heat sink (IHS) and Riding heat sink (RHS) optics. IHS is aimed at current air-cooled data centers, and RHS would be aimed at future liquid‑cooled AI fabrics and extreme port density.
• 7060XE7-64PRS-RV3-L: A specialized 2OU liquid-cooled platform for high-density clusters, featuring 224G SerDes. This system uses DC power from the ORv3 rack and contains no internal fans, integrating with liquid-cooled XPU servers to maximize power efficiency. Available in Q1 2027.
• 7060XE7-128PE: Also coming in Q1 2027, these devices provide 128 800G ports in an air-cooled 4RU design, utilizing 100G SerDes, for environments requiring deployment flexibility and backward compatibility.

On the software side, EOS is the featured network operating system, but the family also supports open-source software such as Software for Open Networking in the Cloud (SONIC) and OpenSwitch. One of the portfolio’s key features is the inclusion of full support for Open Compute Project’s Multipath Reliable Connection (MRC). MRC is an RDMA‑based transport protocol that allows a single reliable connection to simultaneously use many network paths over Ethernet.

“MRC is an open protocol where endstation NICs stripe their traffic across multiple links and paths to the receiver, with out of order packets automatically handled,” wrote Arista’s Kenneth Duda, president and CTO, and Alan Judge, distinguished engineer, in a blog about the technology. “MRC responds to network congestion signals (ECN and packet trimming), shifting load to the best-performing paths, and avoiding links and paths that can’t actually reach the destination altogether.” MRC monitors each path, steering around congestion, avoiding paths with link errors, and avoiding failed links. “We’ve proven in production that this approach achieves very high fabric utilization with good load balancing, while interoperating seamlessly with scale-across and WAN networks utilizing standard dynamic routing protocols,” Duda and Judge wrote.

The software also supports load balancing, congestion management, telemetry and diagnostics, and other technologies that will be core to AI networking. The new Arista family joins a growing ecosystem of vendors looking to tap into the 1.6T Ethernet world, which includes Cisco, Nvidia, Celestica and others. “Arista Network’s new 7060XE7 Series is a strong signal of where large-scale AI fabrics are heading: higher bandwidth, better power efficiency, and tighter integration between compute, optics, silicon, cooling, and network operating software,” wrote Sameh Boujelbene, vice president, data center switch and AI networks market research for Dell Oro, in a LinkedIn post. Among the features that stand out to her are “strong customer and ecosystem validation from Microsoft Azure, Oracle Cloud Infrastructure, Meta, AMD, and Broadcom.”

To provide deeper context, it is worth examining the broader trends driving this launch. The demand for AI infrastructure has skyrocketed with the proliferation of large language models, generative AI, and high-performance computing workloads. These workloads require massive parallel processing across thousands of GPUs or TPUs, which in turn demands extremely high-bandwidth, low-latency networking to avoid bottlenecks. Traditional data center switches operating at 100G or 400G per port are often insufficient for the scale-out fabrics used in AI training clusters. The move to 1.6T per port—offered by Arista's 7060XE7 family—represents a fourfold increase over 400G and a sixteenfold increase over 100G, enabling significantly higher throughput within the same physical footprint.

Broadcom's Tomahawk 6 is a key enabler. This switching silicon provides 51.2 terabits per second of total switching capacity and supports 64 ports of 800G or 256 ports of 200G, among other configurations. By leveraging Tomahawk 6, Arista can offer dense configurations such as 64 ports of 800G in a 4U form factor, or even 128 ports of 800G in a future variant. The support for 224G SerDes in the liquid-cooled model is particularly forward-looking, as it allows for future speed upgrades beyond 800G per port—potentially reaching 1.6T per port in later generations.

The integration with AMD's next-generation compute silicon and NICs is another significant aspect. AMD has been gaining traction in the AI accelerator market with its Instinct series of GPUs. By collaborating with AMD on NICs and compute, Arista is positioning its switches as the preferred fabric for AMD-based AI clusters. This ecosystem approach is essential because AI workloads often require tight coupling between compute, memory, and networking to achieve optimal performance. The support for OCP's MRC protocol further enhances this integration by enabling multipath load balancing and congestion avoidance, which are critical for maintaining high utilization in large-scale fabrics.

Cooling is another major consideration. AI clusters generate enormous heat, and traditional air cooling may not suffice for high-density deployments. The 7060XE7 family offers a range of cooling options: air-cooled for standard data centers, liquid-cooled for extreme density environments, and hybrid designs that combine both. The liquid-cooled 7060XE7-64PRS-RV3-L, for example, uses DC power from the Open Rack v3 standard and has no internal fans, relying on the liquid cooling loop to remove heat. This design can achieve higher power efficiency and support higher port densities than air-cooled counterparts.

The timing of this launch is also notable. Several major cloud providers and enterprises are in the process of upgrading their AI infrastructure to support next-generation models. Microsoft Azure, Oracle Cloud Infrastructure, and Meta have been cited as early supporters of Arista's approach. The competitive landscape includes Cisco's 800G platforms, Nvidia's Spectrum-X Ethernet fabric, and Celestica's white-box switches. However, Arista's strong emphasis on open standards, software quality, and ecosystem collaboration gives it a differentiated position.

In summary, the Arista 7060XE7 Series represents a significant step forward in AI networking. By offering a portfolio that spans air, liquid, and hybrid cooling, supporting the latest Broadcom silicon, embracing open protocols like MRC, and partnering with AMD, Arista is positioning itself as a key enabler of the next wave of AI infrastructure. As AI workloads continue to grow in scale and complexity, the ability to deliver high-bandwidth, low-latency, and power-efficient networking will be critical. The 7060XE7 family appears well-equipped to meet these demands.

Source: Network World News