Pluribus Networks Reveals F64 Series Server-Switch

Pluribus Networks, Inc. announced its F64 Series Server-Switch platform bolstered with Fusion-io, Inc.‘s acceleration technology.

Pluribus Networks was founded with the vision of combining high-performance network, compute and storage together with its Netvisor hypervisor to help network-centric applications move into the network.

By adding Fusion-io to F64 Server-Switch platform, high performance network control and processing are now augmented with in-the-box high-speed storage enabling a new class of applications that can take advantage of these features in the network fabric.

Pluribus’ in-network real-time analytics are now amplified by the ability to store full data flows and sessions in each F64 Server-Switch. Across a fabric of F64 Server-Switches, tens of gigabytes per second of real-time analytics can be captured and processed. Combined with the F64’s fabric-wide multi-million flow processing capability, user applications will have the resources necessary to detect network anomalies and attacks and thwart them more quickly.

Applications running directly in the network which demand high I/O performance and rapid capacity expansion benefit from Fusion-io data acceleration technology without the need for a SAN or NAS. High-performance shared data operations, such as large reduction operations in map-reduce, benefit from the low latency and high bandwidth afforded by hosting data in the network.

"Pluribus Networks is a visionary company that understands the need to virtualize data center networks without compromising performance," said Tyler Smith, Fusion-io VP of Alliances. "With powerful technology from Fusion-io and Pluribus Networks, we believe software-defined solutions will be able to deliver much greater efficiency for customers, enabling them to do more than ever before at a fraction of the cost of yesterday’s network architectures."

The Pluribus Networks Netvisor operating on the F64 Series Server-Switch elevates the network’s programmability and power to match that of modern servers and storage in addition to a high-speed, low-latency network switch and fabric. By optimizing the path between storage and network, customers’ applications reduce the cost and latency associated with unnecessary network hops and bring inline programmability and storage to their data stream. The F64 Series Server-Switch is a 10-Gigabit Data Center Ethernet (DCE) switch that supports industry-standard Layer 2 and Layer 3 protocols as well as customer applications with enterprise server performance in one system.

Key Highlights:
F64 Series Server-Switches and Netvisor Offer Enhanced Control, Visibility and Performance for Latency-Sensitive and Network-centric Applications

• Performance: The F64 Series Server-Switch offers low-latency 10GbE. It supports accurate clocking from external sources as well as an atomic clock option within the device.
• Management: Large fabrics of F64 Series switches can be built and administered from a single point of management as one ‘big logical switch.’ Network capacity is scaled out as workloads and link utilization grow by joining an additional switch to an existing fabric. This means that no manual, switch-by-switch configuration steps have to be performed – traditionally via CLI.
• Greater visibility: Monitor everything. The Netvisor virtualized switch and switch fabric provides a high level of flow capabilities and real-time analytics independent of the end-host server hypervisor environment, whether it is physical or virtual (VMware, Linux KVM, or Xen based). Record packet traffic flows at line rate with Fusion ioMemory.
• Programmable control plane: The Netvisor architecture supports the OpenFlow 1.0 protocol, including in ‘hybrid mode.’ This means that both OpenFlow-controlled and legacy L2/L3 Ethernet networks interoperate in the same environment. A fabric of F64 Series switches can be exposed to OpenFlow controllers as one ‘big logical switch,’ meaning that flow updates are distributed between switches by the Netvisor with high levels of performance at scale.