Infrastructure for the ocr-api service stack.
This consists of one ECS Cluster with one microservice (ocr-api)
These are configured in the profile environmental vars files (no defaults set):
| Variable | Description |
|---|---|
| ec2_instance_type | See [AWS Instance Types)[https://aws.amazon.com/ec2/instance-types/] |
| number_of_tasks | The number of instances of the ocr-api task to run |
| machine_cpu_count | The number of vCPUs the ocr-api uses. |
| machine_amount_of_memory_mib | The amount of memory in MiB to allocate to the ocr-api. |
| ocr_tesseract_thread_pool_size | The number of threads used in the ocr-api application for Tesseract processing (Image to text) |
** Make sure that the CPU and Memory values are in the range of the ec2_instance_type (the plan will be made and applied but fail in deployment with no clear error messages).
Notes on the internal / external naming:
- internal = DNS names within the AWS cluster that are only used for traffic within (internal to) the environment
- external = DNS names exposing specific services to users outside (external to) the cluster e.g. other CH services, testers etc
- so its not a public/private thing its how functionality/services are used - within the cluster is internal DNS, outside users require the external address.
No secrets are required in this application stack
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent