Cisco is boosting network density support for its data center switch and router portfolio as it works to deliver the network infrastructure its customers need for cloud architecture, AI workloads and high-performance computing.
In its data center switch family, Cisco is adding the Nexus 9364E-SG2 switch, a high-density 800G aggregation box that supports numerous port speeds and densities including 400, 200, and 100 Gbps. It also includes support for high-speed optical network connections Open System Form Factor Plus (OSPF) and Quad Small Form Factor Pluggable Double Density (QSPF-DD).
Cisco’s Nexus 9000 data center switches are a core component of the vendor’s enterprise AI offerings. They support innovative congestion-management and flow-control algorithms and deliver the right latency and telemetry to meet the design requirements of AI/ML fabrics, according to Murali Gandluru, vice president of product management with Cisco’s data center networking organization.
The Nexus 9000 portfolio supports critical AI/ML networking features such as dynamic load balancing, which distributes traffic across multiple paths or links that have the same cost in terms of routing metrics, Gandluru wrote in a blog post about the new Nexus 9000 series switches.
Another key feature is priority flow control (PFC), which prevents Ethernet frame drops by signaling, controlling, and managing Ethernet flows along a path by sending pause frames to appropriate senders. In addition, support for explicit congestion notification (ECN) provides end-to-end notification per IP flow by marking packets that experienced congestion, without dropping traffic, Gandluru stated.
Both PFC and ECN are core tenants of the technology that industry group Ultra Ethernet Consortium is developing to enrich the venerable networking technology for AI infrastructure duties.
“Coupled with tools such as Cisco Nexus Dashboard Insights for visibility and Nexus Dashboard Fabric Controller for automation, Cisco Nexus 9000 switches become ideal platforms to build a high-performance AI/ML network fabric,” Gandluru stated.
New router targets high-performance AI clusters
At the higher end of the scale, Cisco rolled out a new high-density Series 8000 router – the 8122-64EH/EHF – that features support for 64 ports of 800G connectivity.
“This new platform is powered by the Cisco Silicon One G200—a 5 nm 51.2T processor that uses 512G x 112G SerDes, which enables extreme scaling capabilities in just a two-rack unit (2RU) form factor,” wrote Guru Shenoy, vice president, product management, in Cisco’s mass-scale infrastructure group, in a blog about the news.
With 64 QSFP-DD800 or OSFP interfaces, the Cisco 8122 supports options for 2x 400G and 8x 100G Ethernet connectivity, Shenoy stated.
All Cisco Silicon processors can be customized for routing or switching from a single chipset, eliminating the need for different silicon architectures for each network function. This is accomplished with a common operating system, P4 programmable forwarding code, and an SDK.
Core to the Silicon One system is its support for enhanced Ethernet features, such as improved flow control, congestion awareness, and avoidance.
The system also includes advanced load-balancing capabilities and “packet-spraying” that spreads traffic across multiple GPUs or switches to avoid congestion and improve latency. Hardware-based link-failure recovery also helps ensure the network operates at peak efficiency, according to Cisco.
Also, like other Silicon One processors, the Cisco 8122 supports open network operating systems, such as Software for Open Networking in the Cloud (SONiC), Shenoy stated.
“Through broad application programming interface (API) support, cloud providers can use tooling for management and visibility to efficiently operate the network. With these customizable options, we are making it easier for hyperscalers and other cloud providers that are adopting the hyperscaler model to meet their requirements,” Shenoy wrote.
In addition to scaling out back-end networks, the Cisco 8122 can also be used for mainstream workloads in front-end networks, such as email and web servers, databases, and other traditional applications, Shenoy stated.
Read more Cisco news
- Cisco revamps key DevNet sandboxes: New and upgraded DevNet sandboxes from Cisco are dedicated to Catalyst Center, CI/CD pipelines for infrastructure automation, and Meraki products.
- Cisco layoffs hit California workers: Cisco’s rolling layoffs are taking a toll on workers in California. Roughly 842 employees impacted by the vendor’s current restructuring plan will come from the San Jose-San Francisco area. There is also speculation – not confirmed by Cisco – that some of its headquarters buildings will be closed.
- Cisco snaps up AI security player Robust Intelligence: Cisco announced plans to acquire Robust Intelligence, a security startup with a platform designed to protect AI models and data throughout the development-to-production lifecycle.
- Cisco adds heft to cybersecurity push with acquisitions, new talent: With new leadership, key acquisitions, and an AI-driven, platform-based vision, Cisco is betting big on security. Its dominance in networking and telecommunications products and services is well established, but its role in cybersecurity is less cemented.
- Cisco debuts CCDE-AI Infrastructure certification: Cisco announced a new certification program it says will empower IT pros to design infrastructure for AI and machine learning. The company’s new Cisco Certified Design Expert (CCDE) AI Infrastructure certification is a vendor-agnostic, expert-level certification.
- Cisco Talos analyzes attack chains, network ransomware tactics: Cisco’s Talos security intelligence group analyzed ransomware groups to identify common techniques and offer recommendations to help security team better protect their businesses.
Source:: Network World