Crush semi-manufacturing runtimes with Calibre in the cloud

The electronic design automation (EDA) software used to manufacture electronic chips is traditionally run on compute clusters on-premises with job schedulers to manage the queues and allocate resources. But cloud is now a viable option for the compute-intensive steps of the post-tapeout (PTFO) flow, including optical proximity correction, mask process correction and mask data preparation. This paper presents solutions that Calibre semiconductor manufacturing PTOF jobs to leverage cloud massive compute power seamlessly and cost-effectively.

We demonstrate following optimizations on an OPC–MPC–MDP PTOF flow launched under Siemens EDA Reference Environment for AWS:

Keep cloud instances busy and provide best runtime by dynamically scaling the jobs and the cluster size; use AWS spot instances seamlessly; monitor jobs and hardware at the same time to fine-tune cloud instance types; reach massive scalability using Calibre FullScale.

Realize scaling and efficiency for the semi-manufacturing flow

The challenges of on-remise compute clusters for the post-tapeout IC flow are reaching a tipping point where adopting cloud solutions becomes more attractive. Implementing a cloud solution for electronic design automation (EDA) workloads is not without it's own challenges, but can be a great solution to many of the problems of managing on-site compute clusters.

The Calibre semi-manufacturing flow is cloud-ready, scales well and is very cost-efficient. Siemens EDA also offers scheduler solutions for the cloud that are well-suited to large post-tapeout flow environments. The Siemens EDA Reference Environment is a pre-configured AWS ParallelCluster. The Calibre Kubernetes Operator launches Calibre jobs on Kubernetes clusters either on-cloud or on-premise.

Cloud computing is a powerful environment to crush Calibre semi-manufacturing runtimes through massive scaling, and Siemens EDA offers the tools to ease the transition to cloud-based post-tapeout flows.