fix: Add documentation to bulkReadRows that each batch will process t…

google-cloud-bigtable/src/main/java/com/google/cloud/bigtable/admin/v2/BigtableTableAdminClient.java

@@ -76,7 +76,7 @@
  * // One instance per application.
  * BigtableTableAdminClient client =  BigtableTableAdminClient.create("[PROJECT]", "[INSTANCE]");
  *
- * CreateTable request =
+ * CreateTableRequest request =
  *   CreateTableRequest.of("my-table")
  *     .addFamily("cf1")
  *     .addFamily("cf2", GCRULES.maxVersions(10))

google-cloud-bigtable/src/main/java/com/google/cloud/bigtable/data/v2/BigtableDataClient.java

@@ -57,7 +57,7 @@
  * // One instance per application.
  * BigtableDataClient client = BigtableDataClient.create("[PROJECT]", "[INSTANCE]")
  *
- * for(Row row : client.readRows(Query.create("[TABLE]")) {
+ * for(Row row : client.readRows(Query.create("[TABLE]"))) {
  *   // Do something with row
  * }
  *
@@ -1077,9 +1077,13 @@ public Batcher<RowMutationEntry, Void> newBulkMutationBatcher(@Nonnull String ta
   }
 
   /**
-   * Reads rows for given tableId in a batch. If the row does not exist, the value will be null.
-   * This operation should be called with in a single thread. The returned Batcher instance is not
-   * threadsafe, it can only be used from single thread.
+   * Reads rows for given tableId in a batch. If the row does not exist, the value will be null. The
+   * returned Batcher instance is not threadsafe, it can only be used from a single thread.
+   *
+   * <p>Performance notice: The ReadRows protocol requires that rows are sent in ascending key
+   * order, which means that the keys are processed sequentially on the server-side, so batching
+   * allows improving throughput but not latency. Lower latencies can be achieved by sending smaller
+   * requests concurrently.
    *
    * <p>Sample Code:
    *
@@ -1113,8 +1117,13 @@ public Batcher<ByteString, Row> newBulkReadRowsBatcher(String tableId) {
 
   /**
    * Reads rows for given tableId and filter criteria in a batch. If the row does not exist, the
-   * value will be null. This operation should be called with in a single thread. The returned
-   * Batcher instance is not threadsafe, it can only be used from single thread.
+   * value will be null. The returned Batcher instance is not threadsafe, it can only be used from a
+   * single thread.
+   *
+   * <p>Performance notice: The ReadRows protocol requires that rows are sent in ascending key
+   * order, which means that the keys are processed sequentially on the server-side, so batching
+   * allows improving throughput but not latency. Lower latencies can be achieved by sending smaller
+   * requests concurrently.
    *
    * <p>Sample Code:
    *