Merge branch 'master' of github.com:aetperf/aetperf.github.io

Author: djfrancesco

_posts/2024-01-08-Streaming-data-from-PostgreSQL-to-a-CSV-file.md

@@ -390,7 +390,7 @@ assert check_df.l_orderkey.is_monotonic_increasing
 
 ## FastBCP
 
-While our focus here is on Python tools, we added FastBCP as a reference regarding CPU and memory usage. FastBCP has been developed in-house by Romain Ferraton at [Architecture & Performance](https://www.architecture-performance.fr/). It is a command line tool, written in C#, that is compatible with any operating system where dotnet is installed. We used dotnet on Linux in the present case.
+While our focus here is on Python tools, we added FastBCP as a reference regarding CPU and memory usage. [FastBCP](https://www.arpe.io/fastbcp) has been developed at [Architecture & Performance](https://www.architecture-performance.fr/ap-logiciels/). It is a command line tool, written in C#, that is compatible with any operating system where dotnet is installed. We used dotnet on Linux in the present case.
 
 FastBCP employs parallel threads, reading data through multiple connections by partitioning SQL on the 'l_orderkey' column, using the "random" method. This approach results in distinct CSV files, later merged into a final output. It's worth mentioning that due to its parallel settings, the resulting data in the CSV file may not be sorted. This is why the ORDER BY clause is removed from the query in this particular case. Also, the returned elapsed time take the merging phase into account.
 

_posts/WP_2024-01-08-Streaming-data-from-PostgreSQL-to-a-CSV-file.md

@@ -370,7 +370,7 @@ assert check_df.l_orderkey.is_monotonic_increasing
 
 ## FastBCP
 
-While our focus here is on Python tools, we added FastBCP as a reference regarding CPU and memory usage. FastBCP has been developed in-house by Romain Ferraton at [Architecture & Performance](https://www.architecture-performance.fr/). It is a command line tool, written in C#, that is compatible with any operating system where dotnet is installed. We used dotnet on Linux in the present case.
+While our focus here is on Python tools, we added FastBCP as a reference regarding CPU and memory usage. [FastBCP](https://www.arpe.io/fastbcp) has been developed at [Architecture & Performance](https://www.architecture-performance.fr/ap-logiciels/). It is a command line tool, written in C#, that is compatible with any operating system where dotnet is installed. We used dotnet on Linux in the present case.
 
 FastBCP employs parallel threads, reading data through multiple connections by partitioning SQL on the 'l_orderkey' column, using the "random" method. This approach results in distinct CSV files, later merged into a final output. It's worth mentioning that due to its parallel settings, the resulting data in the CSV file may not be sorted. This is why the ORDER BY clause is removed from the query in this particular case. Also, the returned elapsed time take the merging phase into account.
 
@@ -426,4 +426,4 @@ While we briefly included FastBCP for a reference comparison, we did not delve i
 
 Additionally, it's worth mentioning that we did not manage to employ [Polars](https://pola.rs/) for the streaming extraction, leading to an out-of-memory error.
 
-Also, it appears that [ConnectorX](https://sfu-db.github.io/connector-x/intro.html) currently lacks support for [retrieving results as Arrow batches or any type of chunks](https://github.com/sfu-db/connector-x/issues/264), making it unfit for this task.
+Also, it appears that [ConnectorX](https://sfu-db.github.io/connector-x/intro.html) currently lacks support for [retrieving results as Arrow batches or any type of chunks](https://github.com/sfu-db/connector-x/issues/264), making it unfit for this task.