andylizf's picture
Upload folder using huggingface_hub
5fed0fc verified
|
raw
history blame
3.1 kB

Cant-Be-Late Multi-Region Scheduling Problem

Multi-region cloud spot instance scheduling benchmark with multiple difficulty variants.

Variants

This problem has 8 variants combining three dimensions:

Dimension Options Description
Availability high_availability East coast regions with good spot availability
low_availability West coast regions with lower spot availability
Deadline tight_deadline 36 hours (12-hour slack)
loose_deadline 48 hours (24-hour slack)
Overhead small_overhead 0.05 hours (3 minutes) restart overhead
large_overhead 0.20 hours (12 minutes) restart overhead

Variant Examples

Variant Deadline Overhead Regions
high_availability_loose_deadline_large_overhead 48h (24h slack) 0.20h (12 min) East coast
high_availability_loose_deadline_small_overhead 48h (24h slack) 0.05h (3 min) East coast
high_availability_tight_deadline_large_overhead 36h (12h slack) 0.20h (12 min) East coast
high_availability_tight_deadline_small_overhead 36h (12h slack) 0.05h (3 min) East coast
low_availability_loose_deadline_large_overhead 48h (24h slack) 0.20h (12 min) West coast
low_availability_loose_deadline_small_overhead 48h (24h slack) 0.05h (3 min) West coast
low_availability_tight_deadline_large_overhead 36h (12h slack) 0.20h (12 min) West coast
low_availability_tight_deadline_small_overhead 36h (12h slack) 0.05h (3 min) West coast

Multi-Region Feature

Unlike single-region cant_be_late, this problem allows:

  • Switching between multiple AWS regions at each timestep
  • Each region has different spot availability patterns
  • Switching regions forces a restart overhead

Evaluation Scenarios

Stage 1: Quick check on 2-region scenario (must pass to proceed)

Stage 2: Full evaluation on multiple scenarios:

  • 2 zones same region (8 traces)
  • 2 regions east-west (8 traces)
  • 3 regions diverse (6 traces)
  • 3 zones same region (6 traces)
  • 5 regions high diversity (4 traces)
  • All 9 regions (2 traces)

Directory Structure

cant_be_late_multi/
├── common/                     # Shared evaluator code
│   ├── __init__.py            # Configuration definitions
│   ├── run_evaluator.py       # Main evaluation logic
│   └── cant-be-late-simulator/  # Simulator with multi-region support
├── {variant}/                  # Each variant directory
│   ├── evaluator.py           # Thin wrapper importing from common/
│   ├── readme                 # Problem specification and API
│   ├── config.yaml            # Variant configuration
│   └── resources/             # Resources for evaluation
└── README.md                  # This file

Common Parameters

All variants share:

  • Task duration: 24 hours
  • On-demand price: ~$3.06/hr
  • Spot price: ~$0.97/hr
  • Time limit: 300 seconds
  • 9 available AWS regions