Workload

OpenDC works with two types of traces that describe the servers that need to be run. Both traces have to be provided as parquet files.

Task

The meta trace provides an overview of the servers:

Metric	Datatype	Unit	Summary
id	string		The id of the server
submission_time	int64	datetime	The submission time of the server
duration	int64	datetime	The finish time of the submission
cpu_count	int32	count	The number of CPUs required to run this server
cpu_capacity	float64	MHz	The amount of CPU required to run this server
mem_capacity	int64	MB	The amount of memory required to run this server

Fragment

The Fragment file provides information about the computational demand of each server over time:

Metric	Datatype	Unit	Summary
id	string		The id of the task
duration	int64	milli seconds	The duration since the last sample
cpu_count	int32	count	The number of cpus required
cpu_usage	float64	MHz	The amount of computational power required.