Table of Contents

Introducing Synopsys Cloud

Cloud native EDA tools and pre-optimized hardware platforms. Experience unlimited EDA licenses with true pay-per-use on an hourly or per-minute basis.

High-performance computing (HPC) is the practice of leveraging multiple computers and storage devices to perform highly complex computations very quickly. An HPC cluster aggregates all of this computational power into a single cohesive framework capable of efficiently running resource-intensive chip design workloads. Traditionally, HPC clusters are built on-premises, but this infrastructure is expensive to build and a hassle to manage. Cloud-based high-performance computing services solve this problem by delivering HPC as a virtualized solution, eliminating the need for chip design firms to build and maintain their own cluster. 


The Benefits of Cloud-Based High-Performance Computing Services for Chip Design

Electronic design automation (EDA) utilizes a lot of processing power, storage, and memory, so it’s a perfect candidate for HPC. EDA workloads and license requirements often fluctuate, sometimes dramatically, which means the underlying HPC infrastructure needs to be highly scalable. The need to protect IP also means that the HPC cluster should be secured by multiple layers of security controls. These factors make cloud-based high-performance computing services the ideal solution for chip design workloads. 

The benefits of cloud-based HPC for chip design include:

 

Easier Scaling

Most semiconductor firms don’t run their HPC clusters at full capacity all the time. There are peaks in utilization during the design phase, then lulls after tape-out or between projects. On-premises HPC clusters aren’t flexible enough to handle these peaks and valleys. There may not be enough resources for complex workflows during the busy season. Still, you’re left running and maintaining expensive computers that you don’t need during the slow periods.

Cloud-based high-performance computing services offer elastic clusters that grow and shrink to meet present demand. Designers always have the resources they need to run EDA workflows, and then the instances are deleted when they’re no longer required. Some cloud-based EDA solutions provide additional flexibility by offering elastic licensing, ensuring designers are never left waiting for an available license and enabling firms to experiment with new tools without an upfront commitment.

 

Lower Costs

An on-premises HPC cluster is expensive to build and maintain. On top of the hardware and software costs, you need to build or rent data center space and hire engineers capable of supporting complex HPC architectures. Scaling an on-premises cluster requires spending even more on hardware, and you have to continue paying for that infrastructure even during lulls when it isn’t being utilized.

Cloud-based HPC services reduce the up-front costs involved in high-performance computing. Vendors distribute the costs of their infrastructure across their entire customer base, reducing the burden on individual firms. Cloud providers are also responsible for supporting and maintaining their own clusters so that semiconductor companies can operate with smaller IT teams. Plus, the elastic scaling of cloud-based HPC means never having to pay for resources that aren’t being utilized. 

 

Improved ROI

Reducing the upfront investment required to adopt HPC can improve the ROI of chip design projects. In addition, cloud-based high-performance computing services allow chip designers to work faster, more efficiently, and with more accurate simulations, reducing time-to-market. The quicker the product gets out the door, the sooner a semiconductor firm can see an ROI. 

 

Reduced Risk

A chip design company’s IP (intellectual property) is how they maintain a competitive advantage, so the security of HPC clusters is critical. However, the complexity of HPC architectures and the need to ensure maximum speed and efficiency make it challenging to keep this infrastructure secure.

In addition, downtime could significantly delay a chip design project and cost thousands of dollars in lost revenue. Building redundant clusters to ensure constant availability means multiplying your infrastructure and staffing costs.

Cloud-based HPC services can help semiconductor firms reduce risk without increasing cost or complexity. Reputable cloud providers use robust security measures to protect their infrastructure and services, including cameras, multi-factor authentication (MFA), and firewalls. Cloud services are also supported by highly specialized teams of HPC and security experts so that chip design companies can focus their IT resources on revenue-generating initiatives. 

Plus, large cloud-based high-performance cloud vendors, like Microsoft Azure, distribute their HPC clusters around the globe for redundancy, which means they’re able to ensure a high degree of reliability.


Synopsys, EDA, and the Cloud

Synopsys is the industry’s largest provider of electronic design automation (EDA) technology used in the design and verification of semiconductor devices, or chips. With Synopsys Cloud, we’re taking EDA to new heights, combining the availability of advanced compute and storage infrastructure with unlimited access to EDA software licenses on-demand so you can focus on what you do best – designing chips, faster. Delivering cloud-native EDA tools and pre-optimized hardware platforms, an extremely flexible business model, and a modern customer experience, Synopsys has reimagined the future of chip design on the cloud, without disrupting proven workflows.

 

Take a Test Drive!

Synopsys technology drives innovations that change how people work and play using high-performance silicon chips. Let Synopsys power your innovation journey with cloud-based EDA tools. Sign up to try Synopsys Cloud for free!


About The Author

Sridhar Panchapakesan is the Senior Director, Cloud Engagements at Synopsys, responsible for enabling customers to successfully adopt cloud solutions for their EDA workflows. He drives cloud-centric initiatives, marketing, and collaboration efforts with foundry partners, cloud vendors and strategic customers at Synopsys. He has 25+ years’ experience in the EDA industry and is especially skilled in managing and driving business-critical engagements at top-tier customers. He has a MBA degree from the Haas School of Business, UC Berkeley and a MSEE from the University of Houston.

Continue Reading