Challenges
A leading provider of engineering systems solutions for the semiconductor and electronics industry wanted to move its fault detection platform to the cloud to reduce costs and accommodate more proofs-of-concept (POCs). The fault detection platform helps manufacturers detect faulty products in near-real-time, identify the root cause of issues, and predict production disruptions. However, with an increasing user base and growing data volume, it was also expensive to maintain and scale a dedicated on-premise environment for POCs.
100 TB of data and growing
Requirement
The solutions provider was looking for a futuristic, scalable, and cost-effective cloud platform that could quickly create environments for POCs and use machine learning and artificial intelligence to predict future defects and take corrective actions. They wanted to operationalize a highly available, scalable, fault-tolerant, and secure data analytics solution for smarter manufacturing. Their future requirements also included cloud enablement programs for internal IT like Dev/QA and customer-facing programs. In addition, they were looking for 8×5 operational support for their cloud infrastructure, data analysis platform, application, and customer tickets.
Deployed a highly secure and fault-tolerant application on AWS
Solution
The Impetus team architected an AWS infrastructure comprising of 10 m4.2xlarge EC2 instances with 3 TB of EBS volumes and 1 TB of RDS storage to support the initial production load. The system was easily scalable to meet the growing demand of users and data volume after moving to production.
After finalizing the design, we provisioned the cloud infrastructure, set up a Cloudera platform, and deployed the application on AWS. Data from the existing on-premise environment was migrated to AWS with near-zero business downtime.
Saved 30% cost on AWS as compared to the on-premise set-up
Impact
Operationalizing a next-generation, cost-effective, fault-tolerant cloud detection platform enabled the client to lower costs, accommodate more use cases, and better manage their growing data volumes.
HIGHLIGHTS
- A hybrid solution (including AWS and on-premise) with dual VPN connectivity
- High availability and fault tolerance using multiple availability zones for EC2 and RDS
- Enterprise-grade security adhering to compliance requirements using AWS KMS and VPC
- End-to-end infrastructure, services, and application monitoring with CloudWatch, Nagios, and custom solutions
- Round-the-clock analysis of cloud resource usage with CloudWatch, AWS Trusted Advisor, and AWS Well-Architected Framework
- A dedicated incident management process to handle customer tickets
