MoSys Announces New Bandwidth Engine Architecture with over 1.5 Tbps Internal Bandwidth
Expanded features, accelerated interfaces and iterative macros increase packet processing capabilities up to 800 Gbps for Next-Generation Networking and Communications Applications
SANTA CLARA, Calif.--Feb. 5, 2014-- MoSys (NASDAQ: MOSY), a leader in semiconductor solutions that enable fast, intelligent data access for network and communications systems, today announced the architecture for its third generation Bandwidth Engine® family, featuring the industry's highest serial throughput and memory access rate, optimized efficiency and scheduling for multi-core processing, and expanded multi-cycle embedded macro functions. The combination of size, access rate and embedded acceleration offloads expands the packet processing performance, addressing the simultaneous increase in both data throughput and network intelligence for next-generation systems targeting networking infrastructure, communications, big data and security applications.
Exponential growth in worldwide network traffic from connected devices and cloud services is creating a need for increasingly complex network management services that support dynamically changing applications and workloads. The third generation Bandwidth Engine architecture delivers unprecedented performance capability, offering over 800 Gigabits per second (Gbps) IO performance, up to 10 billion data accesses per second, and over 1 Gb of high access rate embedded 1T-SRAM® memory. Used in conjunction with FPGA, ASIC or NPU based packet processors, the new Bandwidth Engine architecture accelerates effective throughput and expands lookup and statistics capacity to provide flexible and scalable performance and feature enhancement.
"Our third generation Bandwidth Engine continues our commitment to providing our customers with performance and scalability as well as the fundamental benefits of size, power, pin count and economics," stated John Monson, VP of marketing for MoSys. “The core functionality of this new architecture delivers scale in speed, access, throughput and size, while the integration of multi-cycle offload and threading capabilities enable a new co-processing paradigm for performance and feature expansion."
"Given ever-increasing demand for network bandwidth and new services, system architects are looking to offload network functions, boost concurrent processing capabilities and improve efficiency across the entire network," said Bob Wheeler, principal analyst for networking at The Linley Group. "MoSys’ third generation Bandwidth Engine architecture is driving memory access performance to new heights. The performance of MoSys’ innovative new architecture combined with the efficiency of its open GigaChip Interface protocol supports the access and offload acceleration performance required to meet the needs of 100G, 400G and terabit networks."
MoSys’ new Bandwidth Engine architecture leverages a number of key advancements to enable unprecedented performance. The highly efficient GigaChip™ Interface has been accelerated to support up to 16 lanes running as fast as 28 Gbps following standard OIF CEI electrical specifications. Enhancements have been made to improve efficiency for variable sized commands and packets to maximize effective throughput of data transfer in both directions. For increased size, flexibility and ease of scheduling for multi-core processors, the architecture allows for multiple threads to be issued into extensive multi-cycle macro processing functions or directly across the power and speed optimized 1T-SRAM array. The memory capacity has been scaled to over 1 Gb. The new architecture also supports additional atomic read-modify-write access capabilities to address applications such as coherent multi-threaded access of queues and data structures, as well as n x 100GE network statistics with activity aging, running averages and policing up to 1+ Terabits per second.
Continuing with support for carrier and enterprise class reliability requirements, MoSys’ Bit Safe™ Self-Test and Self-Repair technology provides an additional safety net, maximizing system uptime through a 10 year or longer service life requirement.
Michael Miller, VP of technology innovation and system applications for MoSys , will present additional details on the new third generation Bandwidth Engine architecture in his presentation, “Intelligent Offload Supporting Data Center Equipment at 100G+" at the Linley Tech Data Center Conference 2014 during Session 3, entitled Advancements in Packet Processing, from 3:10-4:40pm on February 5, 2014.
About MoSys, Inc.
MoSys, Inc. (NASDAQ: MOSY) is a fabless semiconductor company enabling leading equipment manufacturers in the networking and communications systems markets to address the continual increase in Internet users, data and services. The company's solutions deliver data path connectivity, speed and intelligence while eliminating data access bottlenecks on line cards and systems scaling from 100G to multi-terabits per second. Engineered and built for high-reliability carrier and enterprise applications, MoSys' Bandwidth Engine® and LineSpeed™ IC product families are based on the company's patented high-performance, high-density intelligent access and high-speed serial interface technology, and utilize the company's highly efficient GigaChip™ Interface. MoSys is headquartered in Santa Clara, California. More information is available at www.mosys.com.
|
Related News
- MoSys Demonstrates 15.625G Bandwidth Engine SerDes Interoperability with Xilinx's Kintex UltraScale FPGA at OFC 2014
- MoSys Unveils New Bandwidth Engine IC with On-Board Macro Functions for 400G Network Equipment
- MoSys Announces Second Generation Bandwidth Engine IC Delivering up to 384 Gbps Throughput in a Single Device
- MoSys Releases Carrier-Grade Bandwidth Engine IC to Production
- MoSys Demonstrates Bandwidth Engine IC Interoperability with LSI SerDes
Breaking News
- TSMC drives A16, 3D process technology
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |