Merge pull request ravendb#1390 from reebhub/RDoc-2005_TCPcompression

RDoc-2005 - TCP Compression page

Documentation/5.3/Raven.Documentation.Pages/server/.docs.json

@@ -46,8 +46,14 @@
     "Mappings": []
   },
   {
-    "Path": "/kb",
-    "Name": "Knowledge Base",
-    "Mappings": []
+      "Path": "/kb",
+      "Name": "Knowledge Base",
+      "Mappings": []
+  },
+  {
+      "Path": "tcp-compression.markdown",
+      "Name": "TCP Compression",
+      "DiscussionId": "48e8e424-a89c-4972-823f-582f11a0c7dd",
+      "Mappings": []
   }
-]
+]

Documentation/5.3/Raven.Documentation.Pages/server/tcp-compression.markdown

@@ -0,0 +1,80 @@
+# TCP Compression
+---
+
+{NOTE: }
+
+* With **TCP Compression** enabled, internal cluster data transfers 
+  are compressed, including nodes' 
+  [database replication](../server/clustering/replication/replication)
+  and the data submitted by 
+  [Data Subscriptions](../client-api/data-subscriptions/what-are-data-subscriptions) 
+  to their workers.  
+
+* Especially on the cloud, the significant reduction in the amount of transferred 
+  data translates to a significant reduction of costs.  
+
+
+* In this page:  
+  * [TCP Compression](../server/tcp-compression#tcp-compression)  
+  * [Compression Algorithm and Ratio](../server/tcp-compression#compression-algorithm-and-ratio)  
+  * [Enabling TCP Compression](../server/tcp-compression#enabling-tcp-compression)  
+
+{NOTE/}
+
+---
+
+{PANEL: TCP Compression}
+
+RavenDB offers a variety of compression tools, including the compression 
+of [stored documents](../server/storage/documents-compression) and 
+[backups](../server/ongoing-tasks/backup-overview#compression).  
+
+With the **TCP compression** feature enabled, data **in transit** is compressed as well.  
+
+* RavenDB will compress data before [replicating](../server/clustering/replication/replication) 
+  it from one cluster node to another.  
+  Replication makes a large portion of a cluster's traffic, and compressing 
+  replicated data will minimize the traffic volume and expedite data delivery.  
+
+* [Data Subscriptions](../client-api/data-subscriptions/what-are-data-subscriptions) 
+  will also transfer compressed data to their workers.  
+  Data subscriptions are used to automate documents processing, 
+  and may transfer large quantities of documents on a regular basis. 
+  Compressing the data they submit can reduce the traffic volume 
+  and the costs of such automated routines considerably.  
+
+## Compression Algorithm and Ratio
+
+* TCP Compression uses the [Zstandard compression algorithm](https://en.wikipedia.org/wiki/Zstandard), 
+  continuously learning your data model to create dictionaries that represent the redundant 
+  structural data across transferred documents.  
+
+* Compression ratio tends to ascend as the size of the transferred data grows, 
+  and may **top 85%** for big data transfers.  
+
+## Enabling TCP Compression
+
+Your server's [license](../studio/server/license-management) 
+type determines whether TCP compression, among other features, 
+is activated on it. No additional configuration is needed 
+to enable this feature.  
+
+![License (Studio View)](images/tcp-compression-license.png "License (Studio View)")
+
+{PANEL/}
+
+## Related Articles
+
+### Studio
+
+- [Documents Compression](../studio/database/settings/documents-compression)  
+- [License](../studio/server/license-management)
+
+### Server
+
+- [Backup Compression](../server/ongoing-tasks/backup-overview#compression)  
+- [Database Replication](../server/clustering/replication/replication)  
+
+### Client API
+
+- [Data Subscriptions](../client-api/data-subscriptions/what-are-data-subscriptions)