eG Monitoring  

Measures reported by EndecaHotSpotTest

This test monitors the performance of specific features, such as clustering, record search, record filter, range filter, content spotlighting, and snippeting to help administrators proactively identify the hotspots and optimize them for improving the overall performance of the server.

Outputs of the test : One set of results the Endeca search application being monitored.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Nav_query_total Indicates the time taken to perform refinements calculations and keyword search functionality combined. Seconds A low value is desired for this measure. Cluster_perf Indicates the time taken for record clustering. Seconds Ideally, the value for this measure should be low. A high value for this measure can cause performance degradation in terms of latency.

The detailed diagnosis of this measure indicates the Metrics time(seconds).
Record_filter_perf Indicates the time taken to perform record filters. Seconds Record filters allows to define the total record sets into subsets and thereby restricting the search and navigation results to these subsets. A low value is desired for this measure.

The detailed diagnosis of this measure indicates the Metrics time(seconds).
Range_filter_perf Indicates the time taken to perform range filters. Seconds Range filters allows the selection of a subset of the entire data set for display and helps navigation queries to return the results from the specific subset of the dataset. Hence, the value of this measure should be low.
Content_spotlight_perf Indicates the time taken to apply Merchandising Rules, including sorting of Merch Rules results. Seconds The time taken for identification and promotion of contextually relevant results from the dataset to the users must be low. A high value can increase the navigation time and causes performance degradation.
Dimension_perf Indicates the time taken for dimension search. Seconds Ideally, the value for this measure should be low.
Spell_engine_perf Indicates the time taken to calculate variations of the spelling and come up DYM or Autocorrect options. Seconds A low value is desired for this measure.
Snippet_Perf Indicates the time taken to find and tag text that matched the keywords. Seconds Ideally, the value for this measure should be low. A high value indicates an increase in the query search time that can cause performance degradation in terms of latency.
Relevance_rank_substring Indicates the time taken for substring. Seconds  
Relevence_rank_exact Indicates the time taken for exacting the records. Seconds Ideally, the value for this measure should be low.
Page_render_total Indicates the time taken to package up the page of results, including aggregation (e.g. applying rollup, paging, and relevance ranking sort). Seconds Ideally, the value of this measure should be low.
Page_render_rcd_list Indicates the the total page render time minus packaging the result into a binary structure. Seconds  
Prefetching_records Indicates the time taken for prefetching horizontal records. Seconds The value for this measure should be low.
Query_result_sort_time Indicates the time taken for sorting the query results. Seconds The value for this measure should be low. A high value for this measure indicates long query processing time.
Record_sort_init_time Indicates the time taken for initialization of record sort. Seconds  
Heap_sort_time Indicates the time taken for sorting heap. Seconds The value for this measure should be low.
Insert_sort_time Indicates the time taken for sorting insertion. Seconds The value for this measure should be low.
Ordinal_insert_sort_time Indicates the time taken for sorting ordinal insertion. Seconds  
Record_ordinal_sort_time Indicates the time taken for sorting record ordinal. Seconds  
Radix_ordinal_sort_time Indicates the time taken for sorting radix ordinal. Seconds  
Sort_time Indicates the time taken for sorting the tie-break. Seconds When two or more dimension values are assigned to the same location by the sort option or sort order then a second sort option and sort order pair can be specified to break the tie. The value for this measure should be low.
Identity_sort_time Indicates the time taken for total identity tie-breaking. Seconds  
Comparison_sort Indicates the time taken for preprocessing comparison sorts. Seconds  
EQL_filter_eval_time Indicates the time taken for EQL stratify filter evaluation. Seconds