Write Time<\/td> 67 min<\/td> 391 min<\/td> 254 min <\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<\/div><\/div>\n\n\n\n
Data Retrieval<\/h2>\n\n\n\n We prepared three test scenarios for data retrieval, each executed on a single thread. Within each scenario, we prepare 100 random queries. To improve result consistency across different data caching systems, each query is repeated, and the execution time of the second query is measured. Execution times are recorded within the scripts performing the tasks. To minimize the influence of external factors on test results, query outputs are not displayed on the screen (although each tool used provides this option to verify result consistency).<\/p>\n\n\n\n
<\/div><\/div>\n\n\n\n
Test Scenarios<\/h2>\n\n\n\n\nA single query consists of randomly selecting one host, one metric, and a 12-hour time range, then retrieving all measurements within that period.* <\/li>\n\n\n\n A single query involves randomly selecting one host and a 12-hour time range, then retrieving all measurements for each metric within that period.*<\/li>\n\n\n\n We randomly select a 12-hour time range and one measurement value. We calculate the average measurement value for hourly intervals across all counters (separately for each counter).<\/li>\n<\/ol>\n\n\n\n* A subset of the TSBS scenarios assumes that for the retrieved counter values, the maximum value will be calculated for each interval. To simulate record retrieval without aggregation, the intervals between measurements were selected so that there is only one value in each interval.<\/p>\n\n\n\n
<\/div><\/div>\n\n\n\n
Test Results<\/h2>\n\n\n\n Table 2: Average execution time of test queries for individual databases.<\/em><\/p>\n\n\n\nScenario Number<\/th> TStorage<\/th> TimescaleDB<\/th> InfluxDB<\/th><\/tr><\/thead> 1<\/td> 1ms<\/td> 135ms<\/td> 2ms <\/td><\/tr> 2<\/td> 1ms <\/td> 176ms <\/td> 6.5ms <\/td><\/tr> 3<\/td> 397ms<\/td> 2090ms <\/td> 4277ms <\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<\/div><\/div>\n\n\n\n
Summary<\/h2>\n\n\n\n In our tests, we compared the performance of three databases in terms of writing and reading records generated by a large number of counters. In both cases, we compared single-node versions writing to a single disk. The tests showed that regardless of whether the user prioritizes data writing or reading, TStorage proved to be the preferred solution for the defined workload scenarios.<\/p>\n\n\n\n
\n
\n
\n
\n
\n
\n
<\/p>","protected":false},"excerpt":{"rendered":"
TStorage \u2013 testy wydajno\u015bciowe Na stronie zawarto opis testu wydajno\u015bci r\u00f3\u017cnych baz danych zaprojektowanych do efektywnego przetwarzania szereg\u00f3w czasowych. W ramach testu por\u00f3wnujemy: Test ma na celu por\u00f3wnanie wydajno\u015bci zapisu, a tak\u017ce odczytu danych. \u0179r\u00f3d\u0142em danych jest symulacja pracy serwer\u00f3w znajduj\u0105cych si\u0119 w centrach danych. Wielko\u015bci\u0105 odczytow\u0105 s\u0105 r\u00f3\u017cne parametry procesora. Test ten jest przygotowany […]<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"bBlocksCursorData":"","advgb_blocks_editor_width":"","advgb_blocks_columns_visual_guide":"","footnotes":""},"class_list":["post-12","page","type-page","status-publish","hentry"],"coauthors":[],"author_meta":{"author_link":"https:\/\/tstorage.eu\/en\/author\/robert-trzcinski\/","display_name":"Robert Trzci\u0144ski"},"relative_dates":{"created":"Posted 1 year ago","modified":"Updated 10 months ago"},"absolute_dates":{"created":"Posted on May 20, 2025","modified":"Updated on August 18, 2025"},"absolute_dates_time":{"created":"Posted on May 20, 2025 1:34 pm","modified":"Updated on August 18, 2025 12:35 pm"},"featured_img_caption":"","featured_img":false,"series_order":"","_links":{"self":[{"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/pages\/12","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/comments?post=12"}],"version-history":[{"count":51,"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/pages\/12\/revisions"}],"predecessor-version":[{"id":912,"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/pages\/12\/revisions\/912"}],"wp:attachment":[{"href":"https:\/\/tstorage.eu\/en\/wp-json\/wp\/v2\/media?parent=12"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/06\/hardware_ts-1024x711.png\")
<\/figure>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/06\/timescale-1-1024x262.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/06\/Influxdb_logo.svg_-2-1024x262.png\"){kind=logo}
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/05\/tstorage_logo-1.png\"){kind=logo}
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/07\/schm1_en.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/07\/schm2_en.png\")
<\/figure>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/07\/schm3_en.png\")
<\/figure>\n<\/div>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/07\/schm4_en.png\")
<\/figure>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/tstorage.eu\/wp-content\/uploads\/2025\/07\/schm5_en.png\")
<\/figure>\n<\/div>\n<\/div>\n\n\n\n