Dynamic performance tuning based on implied data characteristics

One or more non-transitory computer readable media comprising instructions which, when executed by one or more hardware processors, causes performance of operations comprising:

• receiving a set of time series data for a first time interval, the set of time series data comprising a plurality of records associated with a respective plurality of sequential second time intervals, each record comprising a same set of fields including a first field and a second field;
• analyzing the first field across the plurality of records to determine a first number of values stored in the first field across the plurality of records;
• based on the first number being equal to, or less than, a first threshold:
• classifying first data stored in the first field as corresponding to a first data type;
• analyzing the second field across the plurality of records to determine a second number of values stored in the second field across the plurality of records;
• based on the second number exceeding the first threshold:
• classifying second data stored in the second field as a corresponding to a second data type;
• based on classifying the first data as corresponding to the first data type, storing the first data according to a first storage configuration;
• and
• based on classifying the second data as corresponding to the second data type, storing the second data according to a second storage configuration different from the first storage configuration.

A method, comprising:

• training a machine learning model to classify values of time-series data, the training comprising:
• obtaining training data sets, each training data set of historical data comprising:
• time-series data corresponding to a particular interval of time, the time-series data comprising at least one first field across a first plurality of records corresponding to a respective first plurality of intervals of time;
• for each record, a label specifying a particular data type stored in the first field across the first plurality of records, the particular data type associated with a number of different values stored in the first field across the first plurality of records;
• training the machine learning model based on the training data sets;
• receiving a target set of time-series data for a first time interval, the target set of time-series data comprising a second plurality of records associated with a respective plurality of sequential second time intervals, each record comprising a same set of fields including a second field and a third field;
• applying the machine learning model to the target set of time-series data to classify first data stored in the second field across the plurality of records as corresponding to a first data type, and second data stored in the third field across the plurality of records as corresponding to a second data type, the applying comprising:
• analyzing (a) a number of different values stored in the second field across the plurality of records and (b) a number of different values stored in the third field across the plurality of records;
• classifying the first data as corresponding to the first data type based at least on the number of different values stored in the second field across the plurality of records;
• and
• classifying the second data as corresponding to the second data type based at least on the number of different values stored in the third field across the plurality of records.

A method, comprising:

• receiving a set of time series data for a first time interval, the set of time series data comprising a plurality of records associated with a respective plurality of sequential second time intervals, each record comprising a same set of fields including a first field and a second field;
• analyzing the first field across the plurality of records to determine a first number of values stored in the first field across the plurality of records;
• based on the first number being equal to, or less than, a first threshold:
• classifying first data stored in the first field as corresponding to a first data type;
• analyzing the second field across the plurality of records to determine a second number of values stored in the second field across the plurality of records;
• based on the second number exceeding the first threshold:
• classifying second data stored in the second field as a corresponding to a second data type;
• based on classifying the first data as corresponding to the first data type, storing the first data according to a first storage configuration;
• and
• based on classifying the second data as corresponding to the second data type, storing the second data according to a second storage configuration different from the first storage configuration.