From a5a42bbdb0a1fa2e018da5bb255aa0a2ad2c1602 Mon Sep 17 00:00:00 2001
From: Ole Wiedemann <61351036+w-syss@users.noreply.github.com>
Date: Fri, 29 Apr 2022 18:18:01 +0200
Subject: [PATCH] Change "perf" to "performance" (#3841)

The choice of using "perf" instead of "performance" seems a bit out of place, when "perf" is used nowhere else as an abbreviation for "performance".
---
 entity-framework/core/performance/index.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/entity-framework/core/performance/index.md b/entity-framework/core/performance/index.md
index ca2f95b05c..5cdb405e1d 100644
--- a/entity-framework/core/performance/index.md
+++ b/entity-framework/core/performance/index.md
@@ -13,7 +13,7 @@ Database performance is a vast and complex topic, spanning an entire stack of co
 
 As always with performance, it's important not to rush into optimization without data showing a problem; as the great Donald Knuth once said, "Premature optimization is the root of all evil". The [performance diagnosis](xref:core/performance/performance-diagnosis) section discusses various ways to understand where your application is spending time in database logic, and how to pinpoint specific problematic areas. Once a slow query has been identified, solutions can be considered: is your database missing an index? Should you try out other querying patterns?
 
-Always benchmark your code and possible alternatives yourself - the performance diagnosis section contains a sample benchmark with BenchmarkDotNet, which you can use as a template for your own benchmarks. Don't assume that general, public benchmarks apply as-is to your specific use-case; a variety of factors such as database latency, query complexity and actual data amounts in your tables can have a profound effect on which solution is best. For example, many public benchmarks are carried out in ideal networking conditions, where latency to the database is almost zero, and with extremely light queries which hardly require any processing (or disk I/O) on the database side. While these are valuable for comparing the runtime overheads of different data access layers, the differences they reveal usually prove to be negligible in a real-world application, where the database performs actual work and latency to the database is a significant perf factor.
+Always benchmark your code and possible alternatives yourself - the performance diagnosis section contains a sample benchmark with BenchmarkDotNet, which you can use as a template for your own benchmarks. Don't assume that general, public benchmarks apply as-is to your specific use-case; a variety of factors such as database latency, query complexity and actual data amounts in your tables can have a profound effect on which solution is best. For example, many public benchmarks are carried out in ideal networking conditions, where latency to the database is almost zero, and with extremely light queries which hardly require any processing (or disk I/O) on the database side. While these are valuable for comparing the runtime overheads of different data access layers, the differences they reveal usually prove to be negligible in a real-world application, where the database performs actual work and latency to the database is a significant performance factor.
 
 ## Aspects of data access performance