Emergency Response and Wireless Location Systems
by Manlio Allegra
Topics: Main | DSL | VoIP | Wi-Fi | WiMax | Ethernet | 40 Gbps | Financials | Optical Silicon | People | Start-ups | Europe | Japan |
Acquisitions of Silicon Companies | Silicon Directory | Archive
Netronome Unveils 40-Core Network Flow Processors for Unified Computing
Netronome introduced a new line of network flow processors designed for the next phase of convergence in enterprise and carrier data centers -- the unified computing architectures that will combine massive I/O virtualization with high-level content and security packet processing. Netronome's NFP-32xx family addresses this opportunity by combining high-performance network, content and security processing with general purpose processors, such as Intel's IA, through I/O virtualization. Additionally, the NFP-32xx maintains backward-compatibility with the Intel IXP28XX used in a wide range of existing network platforms.
Highlights of the new NFP-32xx network flow processors include:
High-Performance: Up to 40 multi-threaded programmable networking cores running at 1.4GHz enable the NFP to deliver over 56 billion instructions per second with 320 hardware threads that optimize memory utilization. This yields 1800 instructions per packet at 30 million packets per second, enabling 20Gbps of L2-L7 deep packet processing with line-rate security and I/O virtualization for millions of simultaneous flows.
Power Efficiency: Operating at only 15 to 35 watts, the NFP-32xx delivers more than four times the power efficiency of its closest competitor.
Software-Defined I/O: Unlike fixed-function ASICs and configurable NPUs, Netronome's NFPs are fully programmable processors. Software-defined I/O supports both L2-L3 packet processing and L4-L7 application- and content-aware deep packet inspection. This programmability, coupled with line rate packet processing, provides the highest level of inspection and throughput available in the industry for traditional NPU applications such as high-performance line cards.
Network Virtualization in Silicon: Virtualized multicore CPUs are increasingly used in switches, routers, and network appliances for control, data plane, security and other networking applications. To scale network performance to 10Gbps and beyond the NFP-32xx offers a PCIe v2.0 implementation including extended SR-IOV with 256 queues to provide guaranteed bandwidth and low latency access to shared I/O devices. Netronome's SR-IOV drivers include configurable algorithms for load balancing to virtualized multicore CPUs and efficient zero-copy mechanisms that significantly improve CPU, memory and system bandwidth utilization. In addition to eight PCIe lanes, the NFP supports a wide range of popular high-speed network interfaces including dual 25Gbps Interlaken, SPI-4.2 and dual 10Gbps XAUI.
Integrated Hardware Acceleration for Security, Content and Parallel Processing: Unified computing architectures place extensive security requirements on the network. Ranging from coarse policy enforcement using IEEE 802.1AE LinkSec to granular and stateful flow-based applications such as IPsec and SSL, security processing is now required for the majority of network traffic. The NFP provides 20Gbps of line rate security processing by embedding hardware acceleration in a programmable dataplane, including cryptography engines, PKI engines, dedicated hardware for true random number generation, and support for in-line and look-aside security processing designs.
Netronome said it is building a large ecosystem including leading operating system vendors, complementary silicon, protocol software stacks, system design services and other tools vendors. It is also offering development tools derived from the Intel Developer Workbench.
The NFP is available in a range of price and performance options spanning from 16 to 40 cores operating at 1.0 or 1.4GHz with optional cryptography support. The sampling process in underway and general availability is expected in Q3 2009.
http://www.netronome.com
27-May-09
- In November 2007, Netronome Systems, a start-up based in Pittsburgh, Pennsylvania focused on high-performance packet analysis solutions. announced a technology licensing, sales and marketing agreement will Intel focused on its IXP28XX product line of network processors. Under the exclusive deal, Netronome is licensed to continue sales and support of the Intel IXP2805 and the Intel IXP2855 network processors in new designs. In addition, Netronome is developing a next-generation line of IXP-compatible, high-end network flow processors that combine the Intel IXP28XX technology with Netronome's flow management and deep packet inspection architecture.
Search
this site
Search
the Web
Emergency Response and Wireless Location Systems
by Manlio Allegra
Interview: NSN's Sue Spradley on LightSquared's Wholesale Mobile Data Paradigm
by Sue Spradley
Mobile Backhaul Transition Requires New Testing
by Joe Zeto
Cybersecurity Requires Public/Private Coalition
by Greg Oslan
A Rethinking of Cloud Services and Network Architecture
by Jim Theodoras
Delivering Scalable Video Solutions <br><i> Advantages of SVC-Optimized QoS</i>
by Andy Singleton
E-mail Newsletter
