Update performance.md

@ianlewis: fixed according to your proposal

1 files changed, 1 insertions, 1 deletions

diff --git a/g3doc/architecture_guide/performance.md b/g3doc/architecture_guide/performance.md
index ae85f3ddd..b981f0c01 100644
--- a/g3doc/architecture_guide/performance.md
+++ b/g3doc/architecture_guide/performance.md
@@ -211,7 +211,7 @@ url="/performance/applications.csv" title="perf.py http.(node|ruby)
 
 The above figure shows the result of simple `node` and `ruby` web services that
 render a template upon receiving a request. Because these synthetic benchmarks
-do minimal work per request, most like the `redis` case, they suffer from high
+do minimal work per request, much like the `redis` case, they suffer from high
 overheads. In practice, the more work an application does the smaller the impact
 of **structural costs** become.