Software is beginning to take on a bigger role in the SoC design world. How do we get to SW-HW co-verification? This topic was the center of discussion at a private event last week co-located with DVCon. The event, hosted by Jim Hogan and sponsored by Vayavya Labs Pvt. Ltd., included a panel discussion with Frank Schirrmeister (Cadence), Tomas Evensen (Xilinx) and Parag Naik (Saankhya). George Lotridge of VMware and Michael Bair of Intel also gave presentations. Click here for the presentations. (more…)
Posts Tagged ‘www.leepr.com’
What do EDA and IP (as an industry) need to do in 2014 to serve its user base better?
Chris Rowen, Cadence Fellow and Tensilica Founder, will wrap it up with his word on the subject.
“What does the EDA and IP industry need to do in 2014? Simply put, we need to move past EDA.
Let me explain. As an industry, we’re not just about ‘how’ you design something; we’re increasingly about ‘what’ you design.
This comes amid the relentless march of design complexity. It also comes as companies reconsider their position in the electronics ecosystem to try to deliver more value for customers.
For instance, semiconductor vendors are considering where they best fit into the design spectrum and they’re also looking farther upstream to understand market requirements of their customers’ customers. IP providers, for their part, are looking upstream to understand marketing technology requirements better and re-engineering their business models.
“As the new year rolls out, there are promises and associated challenges that the semiconductor industry faces that need attention to ensure the vibrancy of the industry, even as the industry struggles to stay on the Moore’s law trajectory.
First in my list is the area of 2.5D and 3D integration, an area of great promise but with significant challenges. Much has been touted about these approaches as ways to deliver “More than Moore” but it appears to this observer to be advancing at a pace that is slower than hoped for. It seems to be just another year away from full production. But now, enough said, 2014 needs to be the year when much greater focus must be applied to get at least 2.5D technology into mass production. This is not a transitory approach to 3D but rather should last longer in its own right as a very viable technology sitting alongside 3D as 2 approaches to semiconductor integration. 3D still has challenges to be addressed but here again, greater focus needs to be applied to ramp up to full production in 2015.
Next up in our series of predictions Warren Savage, President and CEO of IPextreme, shares with us what he sees in his crystal ball for 2014.
“As the door closes on a successful 2013 for most companies in the semiconductor industry, the outlook for 2014 is bright as we see an explosion of new devices in the so-called “Internet of Things” era. Google’s recent $3.2B acquisition of Nest is indicative that this market will soon eclipse the smart phone/tablet era (aka the post-PC era). The IoT era will bring with it a range of new opportunities for semiconductor companies to exploit that are not mega-devices, but small, specialized technologies that enable opportunities in adjacent markets, like software and data analysis. There may be at least one semiconductor company that exploits this secondary and/or tertiary source of revenue. (more…)
“If 2014 has a watchword for the Semiconductor Industry, it would be momentum and that would be a result of the rapidly increasing use of IP in SoC designs. Add on the mushrooming need for ‘adaptive’ IP to mitigate timing and variation challenges in complex SoCs as performance issues multiply and process geometries shrink.
Moves within the DDR memory space continue to rock the industry and create momentum. Designers are heading directly to the latest JEDEC standard LPDDR4 (low-power DDR4) and moving beyond (or even skipping) LPDDR3 because they’re getting greater gains in performance and low power, an important consideration for mobile applications.
Next up in our series of predictions is the astute insight of Mike Demler, Senior Analyst with The Linley Group & MICROPROCESSOR report, and former EDA & Chip Design news analyst.
“It’s all about the ecosystem triad: EDA + foundry + IP. Cadence and Synopsys continue to evolve more in the IP direction, and there is really not much to say about the tools that hasn’t been said for a long time —just make it all work together! Redundant “standards” and artificial barriers to interoperability cost the semiconductor industry by lowering productivity. This is the problem with the disaggregated model. Back in the days when “real men” had fabs, companies could develop complete design flows without such obstacles.
The triad needs to work together to get over the stall inMoore’s Law at 28nm. Foundries are incurring delays in getting to 16/14nm FinFETS, and almost nobody is going to use 20nm. The chip industry needs an overall lower-cost solution in order to make sub-28nm processes economically viable. Forget 3D ICs, those will be niche products for a long time, about as popular as 3D TV.
Bryon Moyer, Technology Editor at EE Journal, weighs in on what the chip industry needs from EDA and IP in 2014.
“At the low level, this is going to be the year where the push and pull between EDA companies and users determines how easy FinFETs are to design with.
At the high level, it feels to me like IP and EDA need to come closer together. For years, logic design has been done using text because higher-level languages allow better design abstraction and productivity for hand-crafted designs than prior schematic approaches did. When IP entered the scene, designs were mostly hand-done, with occasional bits farmed out to IP. But now IP dominates, whether internal or third-party. Rather than having a custom-logic paradigm that accommodates IP, it feels like we need to move more to an IP paradigm that accommodates custom logic. And it’s not just about logic either: mixed signal is everywhere, and should be included more seamlessly.”
Next up is Brian Bailey, Technology Editor/EDA, for Semiconductor Engineering, who has shared his forecast for EDA in 2014.
“Over the next couple of years there will be an increase in ASIC starts, but not all of these will be for the latest technology nodes or be the massive chips we have come to expect. The new starts will be smaller chips that form the leaves of the Internet of Things, things such as sensors with small amounts of processing and communications. They will be targeting older processes, such as 90nm. Margins on these devices will be slim but volumes high. I see the designers of these products requiring a different mix of capabilities in their EDA tools and different price points. It may create an interesting opportunity for the second tier EDA companies to become significantly bigger.”
“Today the EDA and IP companies alike are in the race to cater to the big semiconductor players by offering the greatest IP in the latest process geometries. However, semiconductor startups, – such as Saankhya – do not change process nodes very quickly because it may not necessarily make business sense. Here is why.
‘Process design margins’ is a fact of life, which we need to add when we use the latest process nodes. Incorporating these design margins simply adds to the overall SoC design cycle time which we want to keep to a minimum. Hence staying with the older process nodes may be a more prudent choice. When we look at the IP domain, we can categorize the available IPs into the following buckets – Analog IPs, Standard Cell and Memory IPs, Processor Soft Cores and Connectivity IPs. Besides the usual PPA (Power, Performance and Area) metrics, what we look at for each of these categories is the metric of Power Efficiency. This metric can be based on mW/Mhz or in some cases mW/sq. mm depending on the kind of tradeoffs which we intend to do at System level.
The following are some of the issues I feel EDA and IP companies should address, especially in the domain of standard cell and memory IPs which are commonly used in almost all ASIC Designs today. (more…)
“Ahhh, the age old question of, ‘What does EDA and IP need to do this year?’ Well, …
The IP folks need a way to provide fully validated IP to their customers with not only a good understanding of all the validation that has been done, but also the expected use model and configurations for the IP. The use model is critical because it is quite common for IP to be ‘abused’ in a way or mode that the IP vendor never expected. When this happens, the customer must be able to assess the problem in-situ and work with the IP vendor to find a resolution. A standardized quality metric is much more desirable than the current option of comparing each vendor’s unique assessment of their validation. This validation and analysis should include, but is not limited to, power, clock domains, testability and physical. The ‘clean’ IP can then be used with more confidence by their customers in their SoC integration.