Maximizing Silicon Design & Chip Performance with Concertio 

Synopsys Editorial Staff

Dec 08, 2021 / 5 min read

As chips get smaller and demand for high-speed everything increases, innovation is no easy feat. The need for real-time and continuous optimization is more important than ever for chipmakers to achieve peak performance and quality from their applications for the entire lifecycle of the product. On top of the pressure on chipmakers to invent more, chips have become inherently more complex, and being able to deploy updates through the design, text, manufacturing, and in-field stages of a chip’s lifespan is not only a desire, but also a necessity. Addressing the need to monitor chip health from cradle to grave, Synopsys developed the SiliconMAX™ Silicon Lifecycle Management (SLM) Platform.

To expand upon the platform’s existing features and add debug/bring-up and in-field operation capabilities, Synopsys recently acquired Concertio Inc., the leading provider of AI-powered performance optimization software. Working cohesively together, Concertio’s software offerings and the Synopsys SLM platform provide deeper visibility into the operation of chips to enable targeted and accurate adjustments. Some of the key applications that would benefit from the optimization features include automotive/industrial IoT devices and cloud-based data centers.

New ElectronicsSemiWikiSemiconductor Engineering, and Electronics Weekly are a few examples of leading industry publications that featured Synopsys’ acquisition of Concertio, highlighting the value of an all-encompassing optimization solution.

As the EDA industry is in a perpetual sprint to evolve, the ability to see a product not just across the finish line of manufacturing, but well into its ongoing use on the field, is vital.

“There is little doubt that Silicon Lifecycle Management is becoming critical to the successful deployment and operation of advanced electronic systems,” said Shankar Krishnamoorthy, general manager, Silicon Realization Group at Synopsys. “This acquisition of Concertio reflects our commitment to aggressively expand the capabilities and benefits of the SiliconMAX SLM platform to address our customers’ rapidly evolving silicon and system health needs.”

SLM Abstract Banner

A Brief History of the SLM Platform

SiliconMAX was first released last year, ushering in a new era of EDA development. The first iteration of the solution provided chipmakers a way to monitor important performance metrics across multiple steps of the silicon lifecycle. The integrated approach includes in-chip monitoring technology and advanced analytics technology. Before adding Concertio, SiliconMAX set its sights on three stages of the IC development process: design, manufacturing, and test.

Silicon Lifecycle Timeline Chart | Synopsys

ith the focus on these three stages, the goal was to target the early steps in the silicon lifecycle where EDA platforms are traditionally used. To address design requirements, we developed our Fusion Design Platform™ with improved silicon-aware design flow. Our Yield Explorer design sensitivity analysis engine enables a quicker production ramp and reduces design-related yield limiters. Lastly, on the testing front, SiliconMAX High-Speed Access IP and TestMAX ALE Adaptive Learning Engine apply adaptive high-bandwidth testing and in-field data monitoring access.

Another principal facet of SiliconMAX is that it employs embedded sensors to determine a variety of factors such as degradation and aging. These sensors and embedded instruments allow manufacturers to view and measure silicon data and test response via stimuli.

“As in any AI/ML application, more data and more analytics foster deeper and better learning. The SiliconMAX SLM Platform performs big data analytics during design, product ramp and manufacturing test. Information from individual die and wafers is uploaded to a unified database over time. When many chips are being manufactured and tested, consolidating and analyzing the data enables ML to spot common issues, identify common solutions, share effective optimizations and identify trends over time,” said Steve Pateras in a SemiEngineering article.

Concertio Steps Up to the Plate

While the original intention of the SiliconMAX platform was primarily to aid in the design and production process, the acquisition of Concertio has allowed us to add even more value beyond design and manufacturing, through the installation of autonomous software agents which act as crucial modes of detection after the chip is in the field. Concertio’s technology also brings to the table artificial intelligence (AI) at the edge with big data analytics in the cloud. The outcome of this integration is the enhanced ability to preventatively self-optimize and regulate systems. Current users of the platform have seen a marked improvement in overall performance of 5-15%.

Concertio Cloud Data Analytics | Synopsys

With Concertio, the platform has firmly established itself as a tool for in-field optimization, giving manufacturers the opportunity to modify the product after it’s been deployed. This type of advanced optimization allows for continuous testing throughout the product’s lifecycle, a feature that is integral to the maintenance of not just the product, but the entire system environment in which it operates.

“For in-field operations, the idea is to observe the software running on the system, analyze it, then tune the system. One example that comes to mind is how a vertically integrated company like Apple has optimized how their MacBook Pro laptop has its battery charged to optimize its lifespan, because they know how often each app is run, what the power and RAM use for an app is, and can then control clock speeds based on workloads, control the RPM rate of fans and ultimately extend the lifetime of the battery,” said Daniel Payne, SemiWiki.

Other important benefits of the new SiliconMAX are data visibility and analysis in real time. If chipmakers can see and compare data from their applications at all times, it opens the door for constant improvement on the original iteration. For example, if a vehicle needed maintenance performed, the manufacturer would be able to predict it ahead of time through models and simulation and perform the necessary alterations to keep the vehicle running at optimal performance.

With enhanced abilities such as embedded agents and a design sensitivity analysis engine, we see a new horizon for chip design. Granting manufacturers the power to see their product through from the early stages of the design process to the end-user deployment has an invaluable impact on the efficiency and success of a project. We look forward to how our users will implement the added capabilities of the SiliconMAX platform and seeing further innovation in EDA.

Continue Reading