IBM introduces advanced design methodology to increase performance and reduce power consumption in custom chips
IBM today introduced an industry-first timing flow created to maximize performance and minimize power consumption in next-generation custom chips.
Called "variation-aware timing," this revolutionary methodology is expected to reduce custom chip design turnaround time by as much as 4X and help to further enable designs that are completed right the first time, meaning faster time-to-market for customers. The methodology is targeted at application specific integrated circuit (ASIC) 130nm (nanometer), 90nm and 65nm chip designs.
"In order to maintain our ASIC leadership, we intend to stay ahead of the innovation curve," said Richard Busch, director of ASIC business development, IBM Systems & Technology Group. "IBM's start-to-finish design solutions seamlessly integrate proven methodologies, tools and services to assure customers fast time to market at a competitive edge. We will continue to invest in leadership tools and methodologies to address the design challenges facing next-generation devices."
As chips become increasingly complex and the industry evolves into using smaller and smaller geometries, there are more variables that could affect the integrity of the design. A slight error can become magnified as dimensions shrink. Not only are the chips smaller, but this closer proximity of devices and wiring causes formerly small variations to now translate into a significant percentage of the electrical signal being transmitted. Factors such as threshold voltages, line width and gate width, all subject to process variation, now significantly impact chip design and performance and must be accounted for in the design methodology.
Using variation-aware timing, IBM engineers are now able to account for the different variables in custom chip design by closely analyzing the time it takes electronic signals to pass from circuit to circuit. With the methodology in place, designers are able to correct errors that might previously have gone undetected, enabling them to maximize performance and minimize power consumption, and ultimately build a more robust chip right the first time.
IBM is able to account for these process and environmental variations in real-time and engineer the chip accordingly throughout the design flow, from development through the manufacturing process.
Variation-aware timing is available now as part of the IBM Cu-11 (130nm) and Cu-08 (90nm) ASIC design kits.
About IBM Microelectronics
IBM is the world's largest information technology company, with 80 years of leadership in helping businesses innovate. IBM is also a recognized innovator in the semiconductor industry, having been first with advances like more power-efficient copper wiring in place of aluminum and faster SOI and silicon germanium transistors. These and other innovations have contributed to IBM's standing as the number one U.S. patent holder for 11 consecutive years. More information about IBM semiconductors can be found at: http://www.ibm.com/chips.
Related Semiconductor IP
- Keccak IP Core
- 9bit time-based two-step ADC - GF 22nm FDX
- 2GS/s 11b dual I/Q DAC on GF 22nm FDX
- JESD204b Deserializer PHY - GF 22nm FDX
- JESD204b Serializer PHY - GF 22nm FDX
Related News
- Dolphin Design chooses DEFACTO's SoC Compiler 9.0: a turnkey methodology to reduce project costs and increase team efficiency
- Ansys and TSMC Collaborate to Deliver Multiphysics Design Methodology for Wireless Chips
- Lightelligence Partners With ZeroPoint Technologies to Increase Data Center Connectivity Performance by Up to 50%
- Harnessing the power of the ecosystem in the era of custom silicon on Arm
Latest News
- Analog Bits Expands Global Engineering Footprint with New Design Center in Dongtan, South Korea
- ChipAgents Helps Whalechip Cut Semiconductor Root Cause Analysis from Days to Minutes
- European Processor Initiative Successfully Concluded
- Kandou AI Selects Baya Systems to Power Its Attack on the AI Memory Wall
- Arteris Expands Partnership with Arm to Accelerate Semiconductor Cybersecurity