WP_Term Object
(
    [term_id] => 16575
    [name] => Rescale
    [slug] => rescale
    [term_group] => 0
    [term_taxonomy_id] => 16575
    [taxonomy] => category
    [description] => 
    [parent] => 386
    [count] => 3
    [filter] => raw
    [cat_ID] => 16575
    [category_count] => 3
    [category_description] => 
    [cat_name] => Rescale
    [category_nicename] => rescale
    [category_parent] => 386
)
            
semiwiki800x100V3 01
WP_Term Object
(
    [term_id] => 16575
    [name] => Rescale
    [slug] => rescale
    [term_group] => 0
    [term_taxonomy_id] => 16575
    [taxonomy] => category
    [description] => 
    [parent] => 386
    [count] => 3
    [filter] => raw
    [cat_ID] => 16575
    [category_count] => 3
    [category_description] => 
    [cat_name] => Rescale
    [category_nicename] => rescale
    [category_parent] => 386
)

Deploying EDA Applications in the Cloud

Deploying EDA Applications in the Cloud
by Kalar Rajendiran on 08-23-2021 at 10:00 am

A company that gets its products to market first stands to gain a competitive edge in the market place. This is even more so in the highly competitive and innovative semiconductor industry. At the same time, designing chips is a very challenging task that involves iterative steps that are computation, memory and storage intensive. EDA tool automation plays a key role in reducing the cycle time to get the product to market.

It’s every company’s dream to have access to infinite compute, memory and storage resources in order to accelerate its product development cycle. From a practical perspective, a commercial cloud platform makes that dream come true. Yet, semiconductor companies were slow to switch to a cloud platform for designing their chips. Primary reasons for this early hesitancy were concern for intellectual property (IP) security and the belief that on-prem resources would be able to handle the job.

But even if the on-prem resources have been optimally planned, all it takes is for one or a few of the many projects to slip their schedules. Suddenly the situation changes from on-prem resources being sufficient to on-cloud computing becoming a necessity.

While a large cross section of the semiconductor industry has since switched to a cloud platform (full cloud or hybrid cloud), there may be a number of companies that are still evaluating this switch. If you’re in this category, a webinar titled “Deploying EDA Applications in the Cloud” would be a good one to watch. It was hosted by Rescale, Inc., a technology company that builds cloud software and services that enables organizations of every size to deliver the engineering and scientific breakthroughs that enrich humanity. Rescale’s mission is to empower anyone to accelerate innovation.

I recently watched the above webinar and the following blog is a synthesis of the salient points I gathered. The webinar flow nicely covers a typical decision-making thought process of “opportunities, challenges, solution and satisfying everyone’s requirements.”

The webinar begins with Jose Fernandez, Principal Semiconductor and Electronics Partnerships, providing an overview of Rescale’s EDA Cloud Platform. He shares five different customer examples where designs were accelerated via the Rescale platform. The design accelerations ranged from savings of 2 days per P&R iteration to 8 weeks quicker time to market (TTM).

Naval Gupte, EDA Solutions Architect picks up from Jose and walks through the opportunities for EDA workflows and challenges in EDA deployments. He then takes us through a demo to highlight how easy and intuitive it is to setup and run jobs in both batch mode and interactive mode via the platform. You will get the full demo experience when you watch the webinar.

Opportunities

EDA workflows demand ability to: scale-out to thousands of small instances, run concurrent independent jobs per user, avoid resource bottlenecks during peak periods, provision large instances with huge memories for long running jobs and support globally distributed teams with highly available and redundant infrastructure spanning multiple geographical regions. In other words, ability to deliver high throughput, handle compute intensive tasks and support global teams 24×7. Cloud platforms are built to provide these capabilities.

Challenges

Typical challenges in an EDA environment are: access to different versions of the same tool, keeping the cost under control, shared access to common data such as libraries and design data, ability to transfer data back and forth at low latencies, ability to integrate with on-prem schedulers and being able to quickly scale-up and scale-out during peak demand periods. Again, cloud platforms are designed and built with these challenges/requirements in mind.

Rescale Solution

A simple user interface with robust automation allows customers to easily setup runs without relying on the IT team. The platform offers lots of templates for running various typical job runs. These templates make it easy to customize for a particular customer’s needs and quickly run jobs.  Rescale can work with its customers to integrate their on-prem job scheduler into the Rescale platform. Rescale makes available, multiple versions of various EDA tools. As long as the customer has the license for a tool, it is highly likely to find on the platform, the version of a tool that customer is looking for.

Satisfying Everyone’s Requirements

An ideal solution addresses the requirements of all stakeholders.  Rescale’s EDA Cloud Platform does exactly that. In essence, the Platform provides an easy to use, powerful solution that supports multi-tool-vendors optimized workflows to run on unlimited resources for faster time to market for their customers’ products. It is a platform that implements multilayer data encryption and multi-factor authentication (MFA) sign-on that has earned Rescale multiple security certifications. The built-in tools make it easy to manage access to software and hardware, enabling easy control of budget.

Refer to figures below.

Rescale EDA Features 1

 

Rescale IT Features 1

Summary

Rescale EDA Cloud Platform offers both breadth and depth. Breadth in terms of not only functionality but also the number of different EDA tools vendors and Cloud vendors supported. Depth in terms of many versions of the tools supported. Rescale offers a secure cloud-based chip design platform that enables a seamless design flow where a customer could tap into one particular set of tools for one chip project and a different set of tools (as per their team needs/skills) for a different chip project. A customer could choose one cloud vendor for one project and could go with a different cloud vendor for another project.

You would want to watch the webinar and follow up with Rescale to explore your own path to deploying EDA applications in the cloud.

 

 

Share this post via:

Comments

There are no comments yet.

You must register or log in to view/post comments.