Acrolinx fix

Author: denrea

articles/cosmos-db/nosql/quickstart-java.md

@@ -54,7 +54,7 @@ You may read more about databases, containers and items [here.](../account-datab
 
 The provisioned throughput is measured in Request Units (*RUs*) which have a monetary price and are a substantial determining factor in the operating cost of the account. Provisioned throughput can be selected at per-container granularity or per-database granularity, however container-level throughput specification is typically preferred. You may read more about throughput provisioning [here.](../set-throughput.md)
 
-As items are inserted into an Azure Cosmos DB container, the database grows horizontally by adding more storage and compute to handle requests. Storage and compute capacity are added in discrete units known as *partitions*, and you must choose one field in your documents to be the partition key which maps each document to a partition. The way partitions are managed is that each partition is assigned a roughly equal slice out of the range of partition key values; therefore you are advised to choose a partition key which is relatively random or evenly-distributed. Otherwise, some partitions will see substantially more requests (*hot partition*) while other partitions see substantially fewer requests (*cold partition*), and this is to be avoided. You may learn more about partitioning [here](../partitioning-overview.md).
+As items are inserted into an Azure Cosmos DB container, the database grows horizontally by adding more storage and compute to handle requests. Storage and compute capacity are added in discrete units known as *partitions*, and you must choose one field in your documents to be the partition key which maps each document to a partition. The way partitions are managed is that each partition is assigned a roughly equal slice out of the range of partition key values; therefore you are advised to choose a partition key which is relatively random or evenly distributed. Otherwise, some partitions will see substantially more requests (*hot partition*) while other partitions see substantially fewer requests (*cold partition*), and this is to be avoided. You may learn more about partitioning [here](../partitioning-overview.md).
 
 ## Create a database account