Merge pull request #9 from IntelLabs/documentation_update

Update README and Install/Run instructions

INSTALL.md

@@ -60,3 +60,20 @@ Flag | Explanation
 --timing    | Compiles server with chronos for internal timing.
 -jX         | Compiles in parallel, using X cores
 INTEL_PATH=path  | Path to the root folder containing pmgd and vcl. Default is "./" which is pmgd and vcl inside vdms folder. Example: scons INTEL_PATH=/opt/intel/
+
+### Running The VDMS Server
+
+The config-vdms.json file contains the configuration of the server.
+Some of the parameters include the TCP port that will be use for incoming
+connections, maximun number of simultaneous clients, and paths to the
+folders where data/metadata will be stored.
+
+**Note:** The folders must already exists in the filesystem.
+
+We provide a script (run_server.sh) that will create some default directories,
+corresponding the default values in the config-vdms.json.
+
+To run the server using the default directories and port, simply run:
+
+    sh run_server.sh
+

README.md

@@ -32,6 +32,15 @@ how it enables efficient vision analytics.
 Here is our [ATC HotStorage '17 Position Paper](https://www.usenix.org/system/files/conference/hotstorage17/hotstorage17-paper-gupta-cledat.pdf).
 Also, take a look at our [presentation](https://www.usenix.org/conference/hotstorage17/program/presentation/gupta-cledat).
 
+## Get Started
+
+To get started, take a look at the [INSTALL.md](INSTALL.md) file, where
+you will find instructions on how to install and run the server.
+
+Also, visit our [wiki](https://github.com/IntelLabs/vdms/wiki)
+to learn more about the VDMS API, and take a look at some of
+the examples/tutorials.
+
 ## Cite
 
 Feel free to include our work into your research!