Transforming HPC Operations: UBLOX Migration to AWS for Enhanced Performance
As the semiconductor industry becomes increasingly competitive, companies must innovate rapidly while managing costs. This case study highlights the challenges faced by a Swiss company specializing in wireless semiconductors, which struggled with high total cost of ownership (TCO), limited scalability and slow go-to-market processes within their High-Performance Computing (HPC) environment. To address these issues, the company sought to migrate to Amazon Web Services (AWS) for a more agile and efficient solution. This case study details the initial challenges, the objectives for migration, the implemented solutions and the benefits achieved, showcasing how cloud adoption transformed their operations and enhanced their ability to innovate.
The Challenge
The customer is a Swiss company that creates wireless semiconductors and modules for consumer, automotive and industrial markets. It operates as a fabless IC and design house. It had an EDA HPC landscape and faced high TCO with inadequate performance. There was a lack of scalability, leading to restricted performance and limited infrastructure visibility.
- Lack of scalability – The customer needed the ability to scale operations and maximize license utilization
- Higher capex investment – The customer wanted to avoid high capex investment
- Slow go-to-market – Accelerating the testing process was crucial to enable faster decision-making by engineers
The Objective
Ensure a reliable, scalable HPC environment for the customer, leading to enhanced overall performance and better TCO while enhancing landscape visibility.
The Solution
It revolves around enhanced performance and built-in automation capabilities to provide a seamless experience and migration of data from on-premises to AWS. The elasticity provided by AWS provided the required scalability and TCO optimization
Assessment
- Conducted a thorough assessment of the current HPC landscape and identified specific bottlenecks
- Identified the specific requirements and constraints of the client's HPC environment
Build
- IBM Spectrum LSF with the Resource Connector feature enabled to dynamically provision Amazon Elastic Compute Cloud (EC2) instances
- Automated provisioning of HPC workload Scheduler
- Auto-scaling of HPC nodes based on the EDA simulation
- Elastic infrastructure that can meet the high resource requirements
- Data migration from on-prem to Amazon FSx for NetApp
Operate
- Seamless user experience for running simulations in the cloud
- End to end day to day operations management
- Continuous monitoring of HPC landscape
The Impact
The migration to AWS helped the customer achieve their HPC transformation goals and achieve faster time-to-value and innovation.
- Enhanced performance with 20 to 30% faster simulation job execution than on-prem
- Cloud-native HPC environment which bursts into the cloud to provide capacity on demand, without changing end-user experience, existing tools or flows
- Efficient and dynamic management of compute hosts and data ensures good performance and optimized costs and lower efforts due to the implementation of automation
- Reduced capex investment
AWS SERVICES
- AWS EC2 with auto scaling
- Amazon Elastic File System (EFS)
- Amazon FSx for NetApp
- Amazon S3
- AWS Identity and Access Management (IAM)
