AWS Infra Optimization - Based on discussion with stakeholder

[emmayank]

Description

Implement a series of AWS infrastructure changes to reduce costs, based on the discussion with Faiz on November 6, 2024. This involves modifying AWS instances, updating infrastructure documentation, and shutting down or consolidating unused resources.

Goals

• [x] Add start and end date for deleted instances in the AWS Master Sheet.
• [x] Explore instance types with lower CPU and higher memory to optimize cost.
• [x] Update the infrastructure landscape diagram and AWS Master Sheet to include CPU/RAM details along with instance names (e.g., t2.large, t3.medium).
• [x] Assess if dsnp-api-frequency-social-web can be merged with an existing instance and shut down if possible.
• [x] For Api-Applications-Prod, stop unused applications and consolidate with existing instances.
• [x] Verify usage of https://api.experience-gpt.becknprotocol.io - retail in reference applications and, if unused, proceed to shut down.
• [x] Obtain approval from Faiz and shut down the beckn_ai instance if authorized.
• [x] Evaluate the possibility of shutting down the MongoDB instance if no longer required.
• [x] Migrate the Jenkins slave into the master instance and shut down the slave.

Expected Outcome

A streamlined AWS infrastructure with optimized costs by eliminating, consolidating, and re-evaluating resource requirements and instance types.

Acceptance Criteria

• [x] Start and end dates for deleted instances are updated in the AWS Master Sheet.
• [x] Low CPU/high memory instance types are evaluated, and applicable instances are identified for replacement.
• [x] Infrastructure landscape diagram and AWS Master Sheet include CPU/RAM details for all instances.
• [x] dsnp-api-frequency-social-web instance is assessed and consolidated or shut down.
• [x] Unused applications on Api-Applications-Prod are stopped, and resources are consolidated with existing instances.
• [x] Usage of https://api.experience-gpt.becknprotocol.io - retail endpoint is verified, and instance is shut down if unnecessary.
• [x] beckn_ai instance is shut down following approval from Faiz.
• [x] MongoDB instance shutdown is evaluated and executed if feasible.
• [x] Jenkins slave instance is migrated to master, and the slave instance is shut down.

Mockups / Wireframes

NA

Product Name

Beckn

Domain

TBD

Tech Skills Needed

• AWS Management
• Cost Optimization
• Infrastructure Documentation
• Jenkins Administration
• MongoDB Management

Complexity

High

Category

Infrastructure Management

Sub Category

AWS Optimization

[prasad-takale-eminds]

Here’s the summary of the activities we’ve completed:

• Added start and end dates to track deleted instances.
• Explored instances with higher memory capacity that could better support our needs.
• Updated CPU and RAM specifications in the infrastructure landscape, including instance names for clarity.
• For Api-Applications-Prod, we stopped unused applications and identified possibilities for consolidating with existing instances.
• Verified the endpoint usage for https://api.experience-gpt.becknprotocol.io - retail within the reference app, as a separate endpoint is designated for the Chat-GPT experience.
• Obtained approval from Faiz to shut down beckn_ai.
• Checked the MongoDB instance to determine if it could be closed – confirmed as closed.
• Permanently deleted instances that were previously shut down.

[prasad-takale-eminds]

Checked the dsnp api frequency supporting application migration possibility. We can't migrate the application from EC2 to other location. Also it is using some block chain code which is using more memory and CPU