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 announcement marks a strategic shift for the company, moving from standalone high-performance switches to complete rack-scale systems that address the extreme density, power, and thermal efficiency demands of modern artificial intelligence workloads.

Overview of the 7060XE7 Series

The 7060XE7 family features fixed switch platforms and configurable rack-scale systems, targeting racks for vertical and horizontal AI workflows. All models 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. The portfolio is built on Broadcom Tomahawk 6 silicon, enabling 1.6T Ethernet speeds per port and massive aggregate throughput.

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. This flexibility is critical as data centers increasingly adopt liquid cooling for high-power GPU clusters.

Product Configurations

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: This is 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. It will be 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.

These configurations demonstrate Arista’s commitment to addressing diverse data center environments, from traditional air-cooled facilities to cutting-edge liquid-cooled deployments. The use of 224G SerDes in the liquid-cooled model signals readiness for next-generation compute architectures.

Software and Protocols

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 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. 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.

The software also supports load balancing, congestion management, telemetry and diagnostics, and other technologies that will be core to AI networking. Arista is also working with AMD on next-generation compute silicon and NICs to enable scale-out AI fabrics, ensuring tight integration between switches and accelerators.

Industry Context and Market Positioning

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. The push to 1.6T is driven by the insatiable bandwidth requirements of large-scale AI training and inference workloads, which often require thousands of GPUs or XPUs to communicate simultaneously. Higher speeds reduce the number of optical cables and switch layers, lowering latency and total cost of ownership.

Industry analysts have noted the significance of the announcement. Sameh Boujelbene, vice president of data center switch and AI networks market research for Dell Oro, stated in a social media post that Arista’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. Among the features that stand out are strong customer and ecosystem validation from Microsoft Azure, Oracle Cloud Infrastructure, Meta, AMD, and Broadcom.

Technical Innovations and Challenges

One of the biggest challenges in AI networking is handling the microbursts and congestion that occur when many nodes send data simultaneously. Traditional Ethernet congestion management mechanisms often fall short. Arista’s implementation of MRC directly addresses this by allowing multipath load balancing at the transport layer, reducing the need for complex per-flow load balancing in the switch fabric. Additionally, the use of Broadcom Tomahawk 6 silicon provides advanced features like in-band telemetry and programmable pipeline stages that can be tailored for AI traffic patterns.

The portfolio also emphasizes thermal management. The liquid-cooled 7060XE7-64PRS-RV3-L is designed to operate without internal fans, reducing power consumption and noise, and enabling deployment in high-density environments where air cooling is insufficient. This aligns with industry trends toward immersion cooling and rear-door heat exchangers for next-generation AI data centers.

Future Outlook

As AI models continue to grow in size—with some exceeding trillions of parameters—the underlying network infrastructure must scale accordingly. Arista’s 7060XE7 Series positions the company to capture a share of this growing market, offering a comprehensive solution that spans from the top-of-rack switch to the fabric spine. With support for open standards like MRC and SONiC, the portfolio also appeals to hyperscale cloud providers who value flexibility and vendor interoperability.

The early availability of air-cooled models in Q4 2026 allows customers to begin testing and deployment immediately, while the liquid-cooled variants arriving in early 2027 provide a path for next-generation facilities. Arista’s collaboration with AMD on compute silicon and NICs further strengthens its ecosystem play, ensuring that the switches can fully exploit the capabilities of future accelerators.

With this launch, Arista demonstrates its ability to innovate at the intersection of networking, cooling, and software, addressing the unique requirements of AI infrastructure. The company’s focus on open protocols and platform flexibility gives customers multiple paths to building high-performance AI fabrics without proprietary lock-in. As the demand for AI compute continues to surge, such solutions will become increasingly critical for enterprises and cloud providers alike.

Source:Network World News