From 905ff7247bff4daa92f85f0060f852e62bd1187e Mon Sep 17 00:00:00 2001 From: Gabor Szarnyas Date: Fri, 6 Dec 2024 17:01:05 +0100 Subject: [PATCH 1/2] caps --- _posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md b/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md index 7bc83aa514..b2012800ea 100644 --- a/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md +++ b/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md @@ -1,6 +1,6 @@ --- layout: post -title: "DuckDB: Running TPC-H SF100 on mobile phones" +title: "DuckDB: Running TPC-H SF100 on Mobile Phones" author: "Gabor Szarnyas, Laurens Kuiper, Hannes Mühleisen" thumb: "/images/blog/thumbs/mobile-benchmarks.svg" image: "/images/blog/thumbs/mobile-benchmarks.png" From 25bf378a95b4161bb0a12e7da3bcaee860f6c122 Mon Sep 17 00:00:00 2001 From: Gabor Szarnyas Date: Fri, 6 Dec 2024 17:06:53 +0100 Subject: [PATCH 2/2] Refine intro and conclusion --- _posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md b/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md index b2012800ea..e485afbadc 100644 --- a/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md +++ b/_posts/2024-12-06-duckdb-tpch-sf100-on-mobile.md @@ -13,7 +13,9 @@ A few weeks ago, we set out to perform a series of experiments to answer two sim 1. Can DuckDB complete the TPC-H queries on the SF100 data set when running on a new smartphone? 2. If so, can a run finish in less than 400 seconds – i.e., faster than the results in the research paper that originally introduced vectorized query processing? -Our quest to answer these took us on an interesting journey. We had a lof of fun and learned the difference between a cold run and a _really cold_ run. Read on to find out. +These questions took us on an interesting quest. +Along the way, we had a lot of fun and learned the difference between a cold run and a _really cold_ run. +Read on to find out more. ## A Song of Dry Ice and Fire @@ -76,7 +78,7 @@ The table contains a summary of the DuckDB benchmark results. ## Historical Context -So why did we embark on the journey of running these experiments in the first place? +So why did we set out to run these these experiments in the first place? Just a few weeks ago, [CWI](https://cwi.nl/), the birthplace of DuckDB, held a ceremony for the [Dijkstra Fellowship](https://www.cwi.nl/en/events/dijkstra-awards/cwi-lectures-dijkstra-fellowship/). The fellowship was awarded to Marcin Żukowski for his pioneering role in the development of database management systems and his successful entrepreneurial career that resulted in systems such as [VectorWise](https://en.wikipedia.org/wiki/Actian_Vector) and [Snowflake](https://en.wikipedia.org/wiki/Snowflake_Inc.). @@ -125,6 +127,7 @@ And here are all results in this post visualized on a plot: />
TPC-H SF100 total query runtimes for MonetDB/X100 and DuckDB
-## Summary +## Conclusion -With our experiments concluded, we can confidently say that, yes, DuckDB can run TPC-H SF100 on a mobile phone and **can even outperform a research prototype running on a high-end workstation in 2004 – with a 2024 smartphone that fits in your pocket.** +It was a long journey from the original vectorized execution paper to running an analytical database on a phone. +But we can keep the conclusion short: yes, DuckDB can run TPC-H SF100 on a mobile phone and **can even outperform a research prototype running on a high-end workstation in 2004 – with a 2024 smartphone that fits in your pocket.**