Cavium Networks (NASDAQ: CAVM), a leading provider of semiconductor products that enable intelligent processing for networking, communications, and the digital home, announced today it has delivered its flagship 32-core OCTEON II CN6880 processor to multiple Tier 1 customers building a wide range of wireless and wired infrastructure networking applications. In OEM benchmarks, the OCTEON II CN6880 provides the industry’s highest performance across many networking, security, deep packet inspection (DPI) and compute intensive solutions based on their own real customer applications. OCTEON II is the first 64 bit processor to provide 48GHz of compute power on a single chip across 32-cores, and provides triple the GHz per chip and 2 to 5x better performance/watt than other available standards-based solutions, with multi-chip system level scalability to up to 1024 cores. OCTEON II’s “Real-Core” architecture also delivers significantly lower-latency and greater determinism than alternative “Pseudo-Core” solutions, which is essential for the increasing multimedia traffic on the Internet, where customers demand fast, responsive processing of real-time voice and video.
The OCTEON II processor family is becoming the leading choice at multiple Tier 1 OEM customers worldwide and has been widely designed into their flagship products across routers, switches, 3G/4G base-stations, services cards, security gateways and appliances, storage networking and mobile infrastructure equipment.
Providers of Next Generation Networks are being challenged to handle the explosive increase in traffic because of the fast adoption of cloud technology, increased exchange of multimedia rich content, and wider deployment of video services by IPTV and OTT (“Over the Top”) providers. Higher traffic coupled with the need for intelligent processing has shifted the bottleneck for L3 – L7 data and security services to CPU processing that requires an unprecedented level of 64bit CPU GHz processing power and dedicated hardware acceleration. Also, the need to support increasing voice and video traffic requires low latency and deterministic processing through dedicated, standards-based cores and an optimized memory subsystem. These applications cannot tolerate the compromises of non-industry-standard CPU cores or non-deterministic, multi-threaded and high-latency cache architectures.
OCTEON II CN6880 marks several breakthrough milestones for standards-based multi-core computing technology. Tier-1 customer boards populated with the CN68XX family are enabling game changing, innovative architectures for next generation networking platforms.
Industry-firsts embodied in the shipping OCTEON II CN6880 32-core single-chip processor include:
- Highest number of standards-based CPU cores in a single chip: 32 MIPS64 cores
- Highest number of dedicated hardware acceleration blocks: Over 100
- Highest compute processing in a single chip: 48GHz of 64 bit processing across 32-cores
- Lowest power consumption per GHz of 64 bit processing
- Highest L2 – L7 Packet Processing including hardware acceleration of packet forwarding, IPsec security, TCP and DPI applications at up to 40+Gbps in a single chip
- Highest Compression/Decompression performance of up to 20+Gbps in a single chip
This level of performance is achieved through OCTEON II’s custom-designed MIPS64 CPU core, superior hardware acceleration, on-chip cross bar interconnect and intelligent load balancing hardware. Customer designs using OCTEON II CN6880 scale performance linearly to large numbers of OCTEON II chips, with actual customer designs using up to 32 OCTEON II processors and up to 1024 cores.
“With its delivery of the CN6880, Cavium Networks has set itself apart and demonstrated its single-chip performance leadership in the market,” said Linley Gwennap, principal analyst of The Linley Group. “These processors, with 4x more cores than standards-based alternatives and unique hardware acceleration for deep-packet-inspection deliver the compute and application performance that OEMs need to keep up with massive increases in network traffic for next-generation cloud, mobile infrastructure, and security applications.”
“Customers have relied on Cavium Networks to deliver breakthrough performance and provide a predictable cadence of doubling the performance/watt every two years,” said Y.J. Kim, General Manager, Infrastructure Processor Group, Cavium Networks. “With the broad shipment of our breakthrough 32-core CN6880, we look forward to extending our leadership in delivering unmatched multi-core solutions for rapid market deployment.”
The Cavium OCTEON processor family is the industry’s broadest and highest performance Multi-core MIPS64 processor family with price points that address application performance ranging from 1Gbps to 40+Gbps in a single chip. The OCTEON processor family consists of processors with 1 to 32 high-performance cores with integrated networking interfaces and industry leading hardware acceleration for packet processing, quality of service, TCP, Security, compression / decompression and DPI.
About Cavium Networks
Cavium Networks is a leading provider of highly integrated semiconductor products that enable intelligent processing in networking, communications and the digital home. Cavium Networks offers a broad portfolio of integrated, software compatible processors ranging in performance from 10 Mbps to 40+ Gbps that enable secure, intelligent functionality in enterprise, data-center, broadband/consumer and access & service provider equipment. Cavium Networks processors are supported by ecosystem partners that provide operating systems, tool support, reference designs and other services. Cavium Networks principal offices are in Mountain View, California with design team locations in California, Massachusetts, India, Taiwan and China. For more information, please visit: http://www.caviumnetworks.com
