The range solves the most complex problems faced by the transition from 10 to 40G and 100G: specifically, high-performance bus routing, on-chip RAM performance and shutdown time of the ultra-high-performance functions required by these systems.

The packet processing solutions, combined with the new Tabula ABAX2P1 3PLD (the first in the ABAX2 P-Series), deliver unique capabilities, including processing four 100G inflows on a single chip, a seeker capable of supporting 100G packet traffic, and a 12x10G-to-100G bridge. These advancements enable the use of Tabula's industry-leading technologies in four key areas: 1) 3D programmable architecture, 2) RTL collector, 3) outstanding frame processing technology, and 4) 3PLD devices. Tabula will demonstrate its high-performance packet processing solutions during the company's inaugural Spacetime Forums series, beginning April 8. This series of one-day technical seminars will run through May in a dozen cities across North America, Asia, and Europe. More than 250 engineers from leading telecommunications and networking OEMs are expected to attend. 

The reference design range for high-performance packet processing consists of:

- A 12x10G-to-100G bridge reference design kit, implementing the aggregation function commonly used in systems of

communications and utilizing the high-performance bus management capabilities of the ABAX2P1 device.

- A 4x100G connection reference design kit, aimed at migrating the data center from 10G to 40G and 100G, made possible by the

ABAX2P1 device capacity to process multiple 100G inputs.

- A second-generation Ternary Search Engine (TSE) reference design kit, which deploys high-performance capacity

search required for featured routers and NGFW while showcasing the unprecedented RAM capabilities of the ABAX2P1 device.

To facilitate user design, the company also deploys a comprehensive set of design examples and IP cores targeting many of the vital high-performance functions available in high-performance packet processing equipment. Examples include a 600Gbps packet classifier, a 100Gbps 64-bit CRC generator, and a 1.3Tbps L2 packet passer.

 

Tabula's four core technology components

Spacetime 3D architecture uses time instead of space as a third dimension, rapidly reprogramming each chip source to perform multiple different functions per cycle—up to 12 within the current generation. Chips using Spacetime, called 3PLDs, have been presented as having three spatial dimensions, with all resources distributed across 12 layers or folds, significantly reducing die size compared to FPGAs. Furthermore, all components in a 3PLD—logic, RAM, multiplier and accumulator blocks, and interconnects—operate in unison at up to 2 GHz, eliminating the performance bottlenecks found in FPGAs.

- The Stylus collector integrates revolutionary technologies such as sequencing time, router-aware placement, and automatic co-optimization of performance and density to deliver a simpler, more intuitive RTL design and a faster loop closure time.

Tabula's partnership with Intel has enabled the manufacturing of ABAX2 P-Series devices using Intel's advanced 22nm Tri-Gate process. These 3D Tri-Gate transistors, the most advanced in the world, deliver unprecedented speed with low operating voltage for reduced power consumption. This key alliance supports production scalability to meet volume demands.

- The ABAX2P1 3PLD is a 12-fold Spacetime device that provides unique factory RAM and logic capabilities alongside IP-directed blocks: a combination that makes the chip and future members of the ABAX2 P-Series perfectly targeted for the most demanding packet processing applications.

By improving all four components, Tabula's high-performance packet processing range provides a programmable solution that delivers industry-leading performance and can be seamlessly deployed even in the most demanding communications and networking roles.

 

More about ABAX2P1

ABAX2P1 3PLD integrates key capabilities that enable system designers to implement high-performance routers, connections, NGFWs, and other communication systems. These include:

- A programmable factory compatible with 2 GHz performance through every chip component - logic, RAM, MAC blocks and interconnects.

- 23.3 MB of 12- and 24-port on-chip memory that delivers 13.8 TB/s of throughput - enough to support multiple 100G inputs.

- Multiple integrated DDR3 controllers operating at 2.133 GT/s, the maximum DDR3 rate, deploying the necessary bandwidth to be compatible with the external buffering or lookup table storage package for multiple entries 100G

-- Multiple integrated 100G Ethernet MACs ensure easy time shutdown and low resource utilization in these ultra-high-performance standard blocks.

Availability

 The initial set of high-performance packet processing solutions is integrated into Stylus and is available now. Additional offerings are planned for monthly release. Designed samples of ABAX2P1 will be available in Q3.

More information or a quote