Big Data ( Cassandra & Spark )

We need to build (Cassandra cluster) & (Apache Spark / Hadoop cluster) in AWS to investigate / POC and Demo to our clients such technology. Build the Cassandra cluster (3 nodes) that can expand as needed scaling the cluster should be matter of minutes with zero or minimum configuration (you need to build the image for the x-node) Build the initial app, this app will create the initial data structure and add the dummy data (Randomly generate by you) Users Table [UserID, Username, FirstName, LastName] Accounts Table (User --[one-to-many]-->Account) [AccountNo, Currency, Balance ] Transaction (Account --[one-to-many]-->Transaction) [TransactionID, time-stamp, details[String, 256 char], category[String, 12 char] ] Build the Spark cluster (same as Cassandra) Update app This app will continuously update Transaction table with dummy transaction data (100-1000 transaction / seconds ) You need to install/configure a driver/pipe to make data available to Spark (from Cassandra) We will need to create a query app (Java + SQL) that will connected and execute some queries (on both Cassandra & Spark) TERMS & CONDITIONS Please bid if you already experienced with (Cassandra / Apache Spark / Hadoop cluster in AWS) Only SQL or Java are allowed (no python or other scripts unless it's used for configuration, i.e. bash scrip is welcome) Only Linux OS. (Documentation + screenshots) of steps taken to create the cluster / scripts and any configuration in nodes/aws We will provide an account in AWS (for that we need some legal document from you , i.e. passport, id , certificate, etc. and signing NDA) The use of "app" does not mean it's a mobile app, it means small application. This will be used by skilled developers, so command line is welcome. (don't waste your time making fancy UI) While in this step Spark is reading data from Cassandra, in the near future Spark should be able to read from other sources and structure and non-structure data (i.e. log files) The idea in this project is to show the speed and scalability of Cassandra / Spark and speed is the key factor for a successful implementation i.e. retrieve the last 3 years transactions in matter of 100-1000 milliseconds This is POC to us and to you (Contractor) so future work could be needed.