Packer

A system for managing and optimizing packet sizes for shipping.

---

Run Locally - Docker Compose Setup

This project includes a Docker Compose configuration to run both the backend and frontend services together.

Prerequisites

• Docker and Docker Compose installed on your system
• Git (to clone the repository)

Running the Application

1. Clone the repository (SSH):

   git clone git@github.com:dsha256/packer.git

2. Navigate to the repository root folder:

   cd packer

3. Start the services using Docker Compose:

   docker-compose up -d

   Or with Taskfile:

   task compose_up

4. Access the application:

   • Frontend: http://localhost:3001
   • Backend API: http://localhost:3000

Stopping the Application

To stop the services:

docker-compose down --remove-orphans

Rebuilding the Services

If you make changes to the code, you'll need to rebuild the services:

docker-compose up -d --build

---

Development

Backend (Go)

The backend is a Go application that provides API endpoints for packet management.

Frontend (Next.js)

The frontend is a Next.js application that provides a user interface for interacting with the backend.

Configuration

The backend configuration is stored in config.yaml. This file is mounted as a volume in the Docker container.

Troubleshooting

If you encounter port conflicts, make sure no other services are using ports 3000 and 3001.

To check the logs:

docker-compose logs -f

To check the logs for a specific service:

docker-compose logs -f backend
docker-compose logs -f frontend

---

Taskfile Commands

This project uses Taskfile for task automation. Here are the available commands:

Development Tasks

• task test - Run all tests with the race flag enabled
• task lint - Run the Go linter to check code quality
• task format - Format all Go code using gofumpt and fieldalignment
• task benchmark_packer - Run the packer service benchmarks

Docker Compose Tasks

• task compose_up - Start the Docker Compose services
• task compose_down - Stop and remove Docker Compose services
• task compose_fresh_restart - Stop all services and restart them with a fresh build

Profiling Tasks

The following tasks launch web UIs for different types of profiling (available at http://localhost:9090):

• task pprof_allocs_web - Memory allocation profiling
• task pprof_heap_web - Heap memory profiling
• task pprof_goroutine_web - Goroutine profiling
• task pprof_block_web - Blocking profiling
• task pprof_threadcreate_web - Thread creation profiling
• task pprof_trace_web - Execution tracing
• task pprof_profile_web - CPU profiling
• task pprof_symbol_web - Symbol lookup

Note: Profiling endpoints are only available when the Go application is running.

---

Algorithms used for packing

Summary of Complexities

Algorithm	Time Complexity	Space Complexity	Approach
CalculateOptimalPacketsForItemsV1	O((items + maxPacketSize) * len(packetSizes))	O(items + maxPacketSize)	Dynamic Programming (Backtracking)
CalculateOptimalPacketsForItemsV2	O((items + maxPacketSize) * len(packetSizes) * log(items + maxPacketSize))	O(items + maxPacketSize)	Dijkstra's Algorithm with Min-Heap

Key Differences in Approach

Factor	V1 (DP)	V2 (Min-Heap / Dijkstra)
Algorithm Type	Dynamic Programming	Priority Queue + Greedy Traversal (Dijkstra)
Main Data Structure	Arrays (dpPacks, prevPacket)	Heap (MinHeap) + Maps (minNumPacks)
Backtracking	Uses prevPacket to reconstruct solution.	Uses predecessor map to reconstruct solution.
Efficiency	Processes all totals up to maxSum.	Prioritizes smaller totals with fewer packets first.

Benchmarks

• Benchmarks and other implementation details are given in this folder internal/packer
• Actual benchmarks' results can be found in this folder benchmark