Chip Design in the Cloud? Now You Can Have Pay-as-You-Go EDA

Article By : Nitin Dahad

Synopsys Cloud will provide access to a suite of cloud-optimized design and verification products with pre-optimized infrastructure on Microsoft Azure, or customers can also bring their own cloud and access the EDA tools.

Synopsys has made a significant move in driving broader adoption of pay-per-use electronic design automation (EDA), with the launch of its pay-as-you-go Synopsys Cloud service. The company said this new cloud-optimized EDA deployment model offers chip and system design flexibility via a single-source, pay-as-you-go approach, helping drive significantly greater productivity and efficiency for increasingly complex chip designs.

This will provide access to a suite of its cloud-optimized design and verification products with pre-optimized infrastructure on Microsoft Azure, or customers can also bring their own cloud and access the EDA tools. Both options are offered in a software-as-a-service (SaaS) model (though pricing options are not available on the company’s web site).

Compared to many other industries, the chip design world has been slow to adopt the cloud – mainly due to concerns over security, as well as other reasons. Writing in a blog, Sandeep Mehndiratta, VP of Synopsys Cloud Solutions, explained, “On-premise cloud, where an organization hosts its own data center infrastructure, does come with challenges in terms of limited compute capacity as well as access to the most advanced compute resources. This model also doesn’t address license management challenges. Tool flows are complicated, encompassing licensing and legal agreements that often involve different vendors with different tool use specifications. Also, there are a fixed number of licenses available, and license servers represent single points of failure.”

This is what Synopsys Cloud intends to address, which the company said is the industry’s first broad-scale cloud SaaS solution for EDA, deployed on the Microsoft Azure cloud computing platform. It aims to provide a single-source, pay-per-use model enabling flexibility for design projects of all sizes by enabling designers to tap into any cloud-optimized tool at any time and on any scale. According to Synopsys, while it offers virtually limitless compute resources and tools, design teams also are freed from the need to manage a complex IT environment. A rich analytics dashboard provides insights on tool usage. And to further simplify the environment, Synopsys said it is also working with major foundries to streamline the foundry collateral process, making it easier for customers to manage foundry data.

Synopsys Cloud graphic
Synopsys said its Synopsys Cloud service is the industry’s first broad-scale cloud SaaS solution for EDA, deployed on the Microsoft Azure cloud computing platform. It aims to provide a single-source, pay-per-use model enabling flexibility for design projects of all sizes by enabling designers to tap into any cloud-optimized tool at any time and on any scale. (Image: Synopsys)

If developers source their own compute infrastructure from public cloud service providers, these ‘bring your own cloud’ or BYOC customers can utilize Synopsys Cloud through a pay-per-use model or even running some workloads in the Synopsys Cloud SaaS environment. Once connectivity has been established between the on-prem and cloud environments to allow data transfer, these customers can simply run the same scripts and binaries for designated design and verification workloads. Usage is measured on an hourly basis and charges are based on activity.

Synopsys said SaaS customers do everything through the Synopsys Cloud portal, logging in to manage users, projects, and infrastructure. Tools for specific use models like timing analysis and physical verification are enabled applications in the SaaS environment, accessible via the portal.

The key benefit for developers is that capacity or hardware constraints don’t need to be an issue, since the new service can help them change the way they manage, schedule, and design EDA projects through the lifecycle of their chips. Mehndiratta comments, “They no longer must commit to a certain number of tool licenses or a particular amount of compute resources before beginning a design. Instead, they can scale up or down as needed, when needed, deciding at any given moment how many instances of a tool are needed and to run them at any scale. As a result, customers can run more jobs faster, taking advantage of the scalability and elasticity of the cloud. Jobs that previously took weeks to finish can now be completed in hours given the dynamic access to robust compute resources.”

Synopsys Cloud BYOC
If developers source their own compute infrastructure from public cloud service providers, these ‘bring your own cloud’ or BYOC customers can utilize Synopsys Cloud through a pay-per-use model or even running some workloads in the Synopsys Cloud SaaS environment. (Image: Synopsys)

Business model: FlexEDA

Synopsys said customers’ most frequent ask was the ability to scale EDA software licenses on demand and with extreme granularity. Vikram Bhatia, responsible for cloud product go-to-market strategy at Synopsys, said, “We wanted to achieve all of this while not making any changes to the EDA tools themselves, so our customers would not have to deal with more versions of the software as they transitioned from on-premises to cloud, either fully or for brief peak usage demands.”

To enable this, the company said its FlexEDA model will provide true pay-per-use and extreme flexibility, potentially changing the way chip design projects are done in the future. The core promise of the FlexEDA model is unlimited EDA license availability with pay-per-use on an hourly or per-minute basis. Chip designers design, schedule, and run a specific EDA job based on where they are in the design cycle and their power, performance, and area (PPA) goals. Synopsys Cloud can automatically scale the EDA software based on the elastic cloud scale infrastructure enabled by the designer for this specific job.

To enable this model, Synopsys Cloud is based on the following architectural elements:

  • A patent-pending metering technology to enable unlimited on-demand EDA tool availability without changing the EDA software code.
  • Pre-optimized compute for each type of EDA workload. This is based on testing and benchmarking hundreds of compute virtual machines available in the public cloud to enable a small subset optimized for specific EDA workloads.
  • Deep integration with the cloud stack to leverage auto-scaling, fault-tolerance, and metering capabilities of the cloud.
  • Browser-based access for end-to-end EDA workload needs. This requires advanced scheduling, virtual desktop, and project and user management capabilities which are native to the platform.
  • In-depth security that elevates the cloud service provider’s shared responsibility model to ensure customer data is secured at all levels with an architecture that fully integrates with all security controls.
  • Ability to leverage FlexEDA in a bring-your-own-cloud (BYOC) deployment model for customers who prefer to run and manage their own public cloud environment.

Addressing both chip complexity and compute trends

Synopsys said chip development in the cloud represents a way forward for an industry grappling with exploding computational demands along with continued time-to-market pressure. From innovative design houses to large systems companies to small startups, more chipmakers are migrating workloads to the cloud to take full advantage of the faster time-to-results, enhanced quality-of-results and better cost-of-results that cloud-based design and verification technologies provide. It has, however, become more challenging to forecast compute needs, leading engineers to underestimate the compute and EDA resources they need while experiencing growing systemic complexity.

“Semiconductor companies are increasingly challenged to quickly deliver both complex functionality and energy efficiency to meet growing requirements for more compute,” said Mark Papermaster, executive vice president and CTO, AMD. “Innovative approaches like Synopsys Cloud built on the Microsoft Azure HBv3 cloud platform – now powered by the latest 3rd Gen AMD EPYC processors with 3D V-Cache technology – can provide access to optimized compute and EDA tools in the cloud, further adding to our innovative chip design capabilities.”

“Our new Synopsys Cloud offering promises to be transformative, providing designers the ability to scale up or down in response to their dynamic chip design and verification needs,” said Sassine Ghazi, president and chief operating officer at Synopsys. “As more design flows incorporate AI, requiring even more resources, the virtually unlimited compute and EDA access we’re providing will lay the foundation for new levels of semiconductor innovation while delivering a flexible, secure chip development environment for future demands.”

Rani Borkar, corporate vice president, Azure hardware systems & infrastructure at Microsoft added, “Addressing systemic complexity along with interdependent design flows in chip design requires more compute and EDA resources than ever before. Microsoft Azure continues to scale its high-performance computing infrastructure with the availability, affordability and capacity to handle advanced chip design and verification workloads. The Synopsys Cloud software-as-a-service solution has been purpose-built on Microsoft Azure for EDA workloads, delivering a flexible design and verification environment to foster the productivity that design teams need.”

A 30-day free trial of Synopsys Cloud is available here.

This article was originally published on Embedded.

Nitin Dahad is a correspondent for EE Times, EE Times Europe and also Editor-in-Chief of embedded.com. With 35 years in the electronics industry, he’s had many different roles: from engineer to journalist, and from entrepreneur to startup mentor and government advisor. He was part of the startup team that launched 32-bit microprocessor company ARC International in the US in the late 1990s and took it public, and co-founder of The Chilli, which influenced much of the tech startup scene in the early 2000s. He’s also worked with many of the big names—including National Semiconductor, GEC Plessey Semiconductors, Dialog Semiconductor and Marconi Instruments.

 

Subscribe to Newsletter

Leave a comment