Avahi Migrates AI Firm to AWS to Scale Delivery of High-Quality 3D Images and Cost-Optimize Cloud Resources

Case Studies >
Avahi Migrates AI Firm to AWS to Scale Delivery of High-Quality 3D Images and Cost-Optimize Cloud Resources

Executive Summary

To ensure machine learning models that turn two-dimensional images into photorealistic, three-dimensional images would scale properly and that cloud compute costs would not spiral out of control, the customer partnered with Avahi Technologies to migrate its infrastructure to Amazon Web Services. Avahi performed a readiness assessment to streamline the migration and allocated tags to compute resources to track costs. Avahi also checked the infrastructure for its ability to provide a secure, high-performing environment that operates efficiently while being resilient and reliable for the customer’s application workloads. By collaborating with Avahi, the customer confirmed the AWS environment is provisioning a cloud infrastructure that’s secure and scalable. The customer also knows they can orchestrate compute resources while controlling costs and providing end-user customers with consistent application performance.

 

Customer Challenge: Scale Machine Learning Models While Controlling Cloud Costs

This start-up firm provides artificial intelligence that allows smartphone users to convert two-dimensional images into three-dimensional lifelike views of objects and scenes. To go to market as quickly as possible, the company deployed an infrastructure-as-code cloud environment, leveraging Kubernetes containers to support machine learning models that enable the high-quality, photorealistic images.

However, the senior leadership team realized that for the 3D service to operate properly, the cloud infrastructure would need to scale on demand. The team was also concerned that without proper management of the infrastructure compute resources, cloud costs would quickly spiral out of control.

 

Partner Solution: Avahi Streamlines Migration and Validates Infrastructure

To ensure the cloud infrastructure would scale while provisioning the 3D service, the firm first decided to migrate the machine learning models to Amazon Web Services (AWS). With data centers available in multiple regions across the globe, AWS can balance workloads and deliver the service close to end-users to decrease latency.

To assist with the migration and validate the cloud architecture would deliver the performance end-users require while also keeping costs in check, the customer partnered with Avahi Technologies. Avahi first performed a readiness assessment as part of the AWS Migration Acceleration Program to streamline the migration from the previous current cloud provider.

Avahi also ran a cost and usage report, using AWS Billing Conductor to ensure all compute resources have properly allocated tags. With an AWS Well-Architected review, Avahi checked the infrastructure for its ability to provide a secure, high-performing environment that operates efficiently while being resilient and reliable for the customer’s application workloads.

Another key tool applied by Avahi was nOps Essential. This allows the customer to cost-optimize and right-size the usage of AWS resources—including storage, IP addresses, virtual private network gateways, load balancers, and servers.

 

Architecture

Results and Benefits: Reduced Migrations Costs and Ongoing Cost Visibility

Avahi validated the customer had chosen wisely in selecting cloud-agnostic tools to build its previous cloud infrastructure. This helped make the migration to the AWS platform a smooth process.

By partnering with Avahi, the customer confirmed the AWS environment is provisioning a cloud infrastructure that’s secure and scalable. The customer also knows they can orchestrate compute resources while controlling costs and providing end-user customers with consistent and resilient application performance.

In addition to assisting with the migration, Avahi identified cost optimization opportunities, which include forming long-term agreements for compute resources. The customer is now pre-buying Amazon EC2 machines for two years rather than deploying servers on demand.

By going through the AWS Well-Architected review with Avahi, the customer became eligible for credits from AWS that reduced the cost of the migration. Just as importantly, the customer has clear visibility into AWS billing and knows exactly how much they pay for every resource, cluster, and operation that span many AWS services including Amazon Elastic Container Service, Amazon Elastic Block Store, and Amazon Aurora . This will allow the customer to properly allocate cloud costs and evaluate the profitability of each end-user customer.

 

About the Partner: Avahi Technologies

Avahi Technologies is a cloud-native focused company and Amazon Web Services (AWS) partner with a team of cloud, data, and software engineering experts and experiences obtained through years of working within the cloud ecosystem. An extraordinary team of highly-certified Avahi experts excels in architecting and operating secure, automated, cloud- based solutions built on AWS. With a focus on becoming an extension to existing customer teams, Avahi offers exceptional service and works tirelessly to build the right solutions to solve business problems.

Confidential
San Francisco Bay Area, CA
Computer Vision, Machine Learning, Augmented Reality
AWS EKS with Karpenter, AWS Aurora, AWS EC2