In order to tune the workloads, we rely on small tweaks.

We should make running and argument passing consistent by :

• Data Caching pass docker_servers.txt for client as an argument instead of modifying it through docker Enhance data caching client #399
• Data Caching separate dataset and warmup to own container Enhance data caching client #399
• Data Caching pure load Enhance data caching client #399
• Data Caching pass argument total number of request to execute instead of infinite Enhance data caching client #399
• Data Serving separate warmup and load to different commands Providing Scripts for Cassandra's Warmup and Test  #398
• Data Serving add command to compact Providing Scripts for Cassandra's Warmup and Test  #398
• Media Streaming the log extraction argument
• Media Streaming should print the connection rate.
• Media Streaming friendly log Error Division by Zero when running cloudsuite media streaming benchmark #386
• Web Search separate dataset and uncompression
• Web Search Simlink instead of copying change dataset copy to symlink #389
• Web Search the tuning parameters should be given to the docker command
• Web Search, and others Add argument to print or not to print log
• Web Serving: delete the tar file after decompression Web Serving Database Now Move the Download Backup #421
• Web Serving: don't copy, use symlink or similar instead Web Serving Database Now Move the Download Backup #421
• Web Serving: merge changes suggested by Mike
• In-memory and graph analytics separate master and worker in documents to match data analytics Analytics workloads upgrade #416
• Fix the memory requirements for analytic workloads 16GB and if possible 32GB
• Add to the configuration parameters of the docker, the number of vcpus that will be allocated to Spark and Hadoop #420
• Django workload (deprecated) Remove Django Workload #397
• For data sets, offer option to either mount from docker image or to mount from folder to control destination directory