Abstract: Methods and systems for ranking analysis results based on user perceived problems of a database system are described. During operation, an embodiment may receive a designation of a perceived database system performance problem from a user, wherein the problem is associated with one or more database system metrics. Next, the embodiment may determine a set of recommendations for alleviating the perceived database system performance problem. The embodiment may then analyze the set of recommendations to determine, for each recommendation in the set of recommendations, an impact the recommendation is expected to have on the one or more database system metrics. Finally, the embodiment may rank the set of recommendations according to the impact each recommendation is expected to have on the perceived database performance problem.

Abstract: Methods and systems for collecting database performance data. During operation a database server can receive a set of T-SQL statements, wherein the database server executes on a computer system and provides access to a database. Next, the database server can execute the set of T-SQL statements, thereby causing the database server to: (1) create an external script which, when executed by the computer system, collects performance data in a file; (2) execute the external script on the computer system; (3) wait for the external script to complete execution; and (4) load the performance data from the file into a table in the database.

Abstract: Methods and systems for adjusting trace filters in a database system are described. During operation, an embodiment may receive, from a user, an indication of a performance problem. Next, the embodiment may determine settings of one or more trace filters based on the indication of the performance problem. The embodiment may then collect trace data based on the one or more trace filters. While the trace data is being collected, the embodiment may perform the following set of operations one or more times: (1) obtain database system performance metrics, (2) analyze the trace data to obtain an analysis result, and (3) adjust the settings of the one or more trace filters based on the database system performance metrics and the analysis result.

Abstract: Methods and systems for identifying one or more indexes are described. A system can collect a set of database operations that were performed during a given time window. Next, the system can partition the set of database operations based on a set of database tables that were accessed by the set of database operations. Then, for each database table in the set of database tables, the system can determine a set of indexes that are candidates for improving database system performance. Next, based on the candidate indexes, the system can identify one or more indexes, one or more combinations of indexes, and/or a consolidated index that improve overall database system performance.

Abstract: Some embodiments provide methods and apparatuses for performing a near-instantaneous restore of a database. During operation, the system intercepts a database management system restore request to restore a database from a backup file to a data file, wherein the backup file comprises a first number (e.g., N1) of data pages. Next, in response to intercepting the database management system restore request, the system sends information to the database management system indicating that the backup file comprises a second number (e.g., N2) of data pages, wherein the second number of data pages is less than the first number of data pages (e.g., N2<N1). Once the database management system has copied the fewer number of data pages from the backup file to the data file, the database management system may indicate that the restore is complete, and continue processing database requests as if the database management system was back online.

Abstract: Methods and systems for adjusting trace filters in a database system are described. During operation, an embodiment may receive, from a user, an indication of a performance problem. Next, the embodiment may determine settings of one or more trace filters based on the indication of the performance problem. The embodiment may then collect trace data based on the one or more trace filters. While the trace data is being collected, the embodiment may perform the following set of operations one or more times: (1) obtain database system performance metrics, (2) analyze the trace data to obtain an analysis result, and (3) adjust the settings of the one or more trace filters based on the database system performance metrics and the analysis result.

Abstract: Methods and systems for identifying one or more indexes are described. A system can collect a set of database operations that were performed during a given time window. Next, the system can partition the set of database operations based on a set of database tables that were accessed by the set of database operations. Then, for each database table in the set of database tables, the system can determine a set of indexes that are candidates for improving database system performance. Next, based on the candidate indexes, the system can identify one or more indexes, one or more combinations of indexes, and/or a consolidated index that improve overall database system performance.

Abstract: Methods and systems for collecting performance data are described. During operation a database server can receive a set of T-SQL statements, wherein the database server executes on a computer system and provides access to a database. Next, the database server can execute the set of T-SQL statements, thereby causing the database server to: (1) create an external script which, when executed by the computer system, collects performance data in a file; (2) execute the external script on the computer system; (3) wait for the external script to complete execution; and (4) load the performance data from the file into a table in the database.

Abstract: Methods and systems for ranking analysis results based on user perceived problems of a database system are described. During operation, an embodiment may receive a designation of a perceived database system performance problem from a user, wherein the problem is associated with one or more database system metrics. Next, the embodiment may determine a set of recommendations for alleviating the perceived database system performance problem. The embodiment may then analyze the set of recommendations to determine, for each recommendation in the set of recommendations, an impact the recommendation is expected to have on the one or more database system metrics. Finally, the embodiment may rank the set of recommendations according to the impact each recommendation is expected to have on the perceived database performance problem.

Abstract: Some embodiments provide methods and apparatuses for performing a near-instantaneous restore of a database. During operation, the system intercepts a database management system restore request to restore a database from a backup file to a data file, wherein the backup file comprises a first number (e.g., N1) of data pages. Next, in response to intercepting the database management system restore request, the system sends information to the database management system indicating that the backup file comprises a second number (e.g., N2) of data pages, wherein the second number of data pages is less than the first number of data pages (e.g., N2<N1). Once the database management system has copied the fewer number of data pages from the backup file to the data file, the database management system may indicate that the restore is complete, and continue processing database requests as if the database management system was back online.

Abstract: There is a variety of media that may be inserted into a reading or writing device, such as CD's, USB drives, floppy disks, memory sticks, and many other devices. Media is inserted into a media reading or writing device that is in communication with a computer or network device. Upon insertion of the media, a number of metadata regarding that media is available to the computer. The trustworthy calculator is typically a plug-in software module that processes each piece of volume metadata and applies a weighed score, resulting in a Trustworthy Factor. A scoring matrix denotes ranges of values of the Trustworthy Factor into a Level of Trust Zone. Based on the Level of Trust Zone, appropriate action handlers may direct the computer to disallow the mounting of the media, may require specific authentication action to take place prior to allowing a mount of the media, or may indicate that the media may be mounted without further authentication.

Abstract: There is a variety of media that may be inserted into a reading or writing device, such as CD's, USB drives, floppy disks, memory sticks, and many other devices. Media is inserted into a media reading or writing device that is in communication with a computer or network device. Upon insertion of the media, a number of metadata regarding that media is available to the computer. The trustworthy calculator is typically a plug-in software module that processes each piece of volume metadata and applies a weighed score, resulting in a Trustworthy Factor. A scoring matrix denotes ranges of values of the Trustworthy Factor into a Level of Trust Zone. Based on the Level of Trust Zone, appropriate action handlers may direct the computer to disallow the mounting of the media, may require specific authentication action to take place prior to allowing a mount of the media, or may indicate that the media may be mounted without further authentication.