{"data_set_description":{"id":"562","name":"cpu_small","version":"2","description":"**Author**:   \n**Source**: Unknown - Date unknown  \n**Please cite**:   \n\nThe Computer Activity databases are a collection of computer systems\nactivity measures. The data was collected from a Sun Sparcstation\n20\/712 with 128 Mbytes of memory running in a multi-user university\ndepartment. Users would typically be doing a large variety of tasks\nranging from accessing the internet, editing files or running very\ncpu-bound programs.  The data was collected continuously on two\nseparate occasions. On both occassions, system activity was gathered\nevery 5 seconds. The final dataset is taken from both occasions with\nequal numbers of observations coming from each collection epoch.\n\nSystem measures used:\n1. lread - Reads (transfers per second ) between system memory and user memory.\n2. lwrite - writes (transfers per second) between system memory and user memory.\n3. scall - Number of system calls of all types per second.\n4. sread - Number of system read calls per second.\n5. swrite - Number of system write calls per second .\n6. fork - Number of system fork calls per second.\n7. exec - Number of system exec calls per second.\n8. rchar - Number of characters transferred per second by system read calls.\n9. wchar - Number of characters transfreed per second by system write calls.\n10. pgout - Number of page out requests per second.\n11. ppgout - Number of pages, paged out per second.\n12. pgfree - Number of pages per second placed on the free list.\n13. pgscan - Number of pages checked if they can be freed per second.\n14. atch - Number of page attaches (satisfying a page fault by reclaiming a page in memory) per second.\n15. pgin - Number of page-in requests per second.\n16. ppgin - Number of pages paged in per second.\n17. pflt - Number of page faults caused by protection errors (copy-on-writes).\n18. vflt - Number of page faults caused by address translation.\n19. runqsz - Process run queue size.\n20. freemem - Number of memory pages available to user processes.\n21. freeswap - Number of disk blocks available for page swapping.\n22. usr - Portion of time (%) that cpus run in user mode.\n23. sys - Portion of time (%) that cpus run in system mode.\n24. wio - Portion of time (%) that cpus are idle waiting for block IO.\n25. idle - Portion of time (%) that cpus are otherwise idle.\n\nThe two different regression tasks obtained from these databases are:\n\nCompAct\nPredict usr, the portion of time that cpus run in user mode from all attributes 1-21.\n\nCompAct(s)\nPredict usr using a restricted number (excluding the paging information (10-18)\n\nOriginal source: DELVE repository of data.\nSource: collection of regression datasets by Luis Torgo (ltorgo@ncc.up.pt) at\nhttp:\/\/www.ncc.up.pt\/~ltorgo\/Regression\/DataSets.html\nCharacteristics: 8192 cases, 13 continuous attributes","description_version":"1","format":"ARFF","creator":"Luis Torgo","upload_date":"2014-10-03T21:52:50","language":"English","licence":"Public","url":"https:\/\/openml.org\/data\/v1\/download\/52740\/cpu_small.arff","parquet_url":"https:\/\/data.openml.org\/datasets\/0000\/0562\/dataset_562.pq","file_id":"52740","default_target_attribute":"usr","visibility":"public","original_data_url":"http:\/\/www.ncc.up.pt\/~ltorgo\/Regression\/DataSets.html","status":"active","processing_date":"2020-11-20 20:18:37","md5_checksum":"5a6047bac59358cc7b8edd507165f8bc"}}