Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/cadence-collaborates-with-memverge-to-increase-resiliency-and-cost-optimization-of-long-running-high-memory-eda-jobs-on-aws-spot-instances.20066/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021370
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Cadence Collaborates with MemVerge to Increase Resiliency and Cost-Optimization of Long-Running High-Memory EDA Jobs on AWS Spot Instances

AmandaK

Administrator
Staff member
17 Apr 2024

Building upon its 20+ years of leadership in providing hosted (or managed) cloud solutions for EDA workloads, Cadence has collaborated with MemVerge to enable seamless support for AWS Spot instances for long-running high-memory EDA jobs.

Cost-optimization of cloud infrastructure has become a consistent ask as more and more customers deploy EDA workflows on Cadence Cloud. Utilizing AWS Spot instances, which provide up to 90% cost savings over on-demand pricing, is one of the most effective ways to save costs on the cloud. However, not all EDA jobs can take advantage of Spot instances.

EDA jobs, specifically in the back-end design workflow, can take several days to complete and cannot take advantage of Spot instances because they can be taken away within two minutes of notification. Hence, the job running on the Spot instance must be cold restarted, resulting in a loss of engineering productivity and resource wastage of several runtime hours and compute costs.

In a move that promises significant cost savings and enhanced efficiency for design engineers, the Cadence and MemVerge collaboration solves this challenge by implementing a transparent, low-overhead incremental checkpoint/restore solution that makes these EDA jobs resilient (hot restart) to Spot pre-emptions or without needing to change the underlying EDA application.

Key customer benefits include:
  • - Cost Savings: By utilizing Spot instances, users can enjoy significant cost savings compared to traditional on-demand pricing models without compromising performance or reliability.
  • - No Application Modification: The integration between Cadence’s EDA solutions and MemVerge's MMCloud technology ensures a seamless user experience, with no need to modify existing EDA applications.
  • - Resiliency and Reliability: Incremental checkpoint and restore technology provides users with the ability to achieve deterministic results, ensuring reliability and consistency even in the event of system failures or interruptions, whether on-prem or on AWS.
For complex advanced-node chip designs, the Cadence Innovus™ Implementation System can run for several days. By adopting this incremental checkpoint solution, even large memory jobs using the Innovus system can run without disruption. Based on real-world production test case modeling, Innovus users can realize up to 57% and 48% cost savings over on-demand pricing for design top and design block jobs, respectively.

Cadence is committed to delivering innovative solutions that enable customers to run EDA workflows in the cloud effectively and cost-efficiently, thus increasing developer productivity.

For more information about this collaboration and how it can benefit your EDA workflows, please see this blog post on how Innovus is now Spot Instance Ready(opens in a new tab).

Get more insights in this AWS blog(opens in a new tab) post on how Spot instances can provide up to 90% cost savings for long-running HPC jobs.

View demos of Cadence tools, such as the Innovus system, running with the MemVerge solution.

Learn more about Cadence OnCloud solutions and sign up for a free trial.

Link to Press Release
 
Back
Top