Apply suggestions from code review

Co-authored-by: markzegarelli <[email protected]>

Author: stayseesong

src/connections/storage/warehouses/choose-warehouse.md

@@ -17,13 +17,13 @@ Both Redshift and BigQuery are attractive cloud-hosted, affordable, and performa
 
 ## Architecture
 
-When you provision a Redshift cluster, you're renting a server from Amazon Web Services. Your cluster comprises of [nodes](http://docs.aws.amazon.com/redshift/latest/dg/c_high_level_system_architecture.html), each with dedicated memory, CPU, and disk storage. These nodes handle data storage, query execution, and - if your cluster contains multiple nodes - a leader node will handle coordination across the cluster.
+When you provision a Redshift cluster, you're renting a server from Amazon Web Services. Your cluster consists of [nodes](http://docs.aws.amazon.com/redshift/latest/dg/c_high_level_system_architecture.html), each with dedicated memory, CPU, and disk storage. These nodes handle data storage, query execution, and - if your cluster contains multiple nodes - a leader node will handle coordination across the cluster.
 
 Redshift performance and storage capacity is a function of cluster size and cluster type. As your storage or performance requirements change, you can scale up or down your cluster as needed.
 
 With BigQuery, you're not constrained by the storage capacity or compute resources of a given cluster. Instead, you can load large amounts of data into BigQuery without running out of memory, and execute complex queries without maxing out CPU.
 
-This is possible because BigQuery takes advantage of distributed storage and networking to separate data storage from compute power. Data distributes across many servers in the Google cloud using their [Colossus distributed file system](https://cloud.google.com/blog/big-data/2016/01/bigquery-under-the-hood). When you execute a query, the [Dremel query engine](https://cloud.google.com/blog/big-data/2016/01/bigquery-under-the-hood) splits the query into smaller sub-tasks, distributes the sub-tasks to computers across Google data centers, and then re-assembles them into your results.
+This is possible because BigQuery takes advantage of distributed storage and networking to separate data storage from compute power. Google's[Colossus distributed file system](https://cloud.google.com/blog/big-data/2016/01/bigquery-under-the-hood) distributes data across many servers in the Google cloud. When you execute a query, the [Dremel query engine](https://cloud.google.com/blog/big-data/2016/01/bigquery-under-the-hood) splits the query into smaller sub-tasks, distributes the sub-tasks to computers across Google data centers, and then re-assembles them into your results.
 
 ## Pricing
 

src/connections/storage/warehouses/redshift-tuning.md

@@ -13,7 +13,7 @@ To help you improve your query performance, this guide takes you through common
 
 As your data volume grows and your team writes more queries, you might be running out of space in your cluster.
 
-To check if you're getting close to your max, run this query. It will tell you the percentage of storage used in your cluster. Segment recommends never exceeding 75-80% of your storage capacity. If you're nearing capacity, consider adding some more nodes.
+To check if you're getting close to your max, run this query. It will tell you the percentage of storage used in your cluster. Segment recommends that you don't exceed 75-80% of your storage capacity. If you approach that limit, consider adding more nodes to your cluster.
 
 ![](images/asset_HvZs8FpE.png)
 
@@ -61,7 +61,7 @@ As mentioned before, Redshift schedules and prioritizes queries using [Workload
 
 The default configuration is a single queue with only 5 queries running concurrently, but Segment discovered that the default only works well for low-volume warehouses. More often than not, adjusting this configuration can improve your sync times.
 
-Before Segment's SQL statements, Segment uses `set query_group to "segment";` to group all the queries together. This allows you to create a queue just for Segment that isolates from your own queries. The maximum concurrency that Redshift supports is 50 across _all_ query groups, and resources like memory distribute evenly across all those queries.
+Before Segment's SQL statements, Segment uses `set query_group to "segment";` to group all the queries together. This allows you to create a queue that isolates Segment's queries from your own. The maximum concurrency that Redshift supports is 50 across _all_ query groups, and resources like memory distribute evenly across all those queries.
 
 Segment's initial recommendation is for 2 WLM queues: