Through an existing customer recommendation, a client came to us with a fairly common request; they wanted to increase the speed of geological calculations, processing existing well data, as well as exploratory data.
Whilst previously we would have had to look at teaming, using multiple processors and memory from workstations on the network or buy in servers to do the task, by utilising the cloud we setup a RDS server in Amazon AWS.
We used a c4.8xlarge instance with 36 CPUs and 60GB Ram, this compute-optimised instance was able to dramatically cut down the processing times as requested. Times were cut down to less than a quarter which meant data was easier and quicker at hand when required.
This instance was more than capable for the task required from it, however the other advantage we have with AWS is that our customer is only charged for when it is on. To take full advantage of this we have implemented scripts to shut down the server when the CPU is idle for more than 2 hours and also scripts to enable the end users to start up the servers when required.
Given the choice of options for our client, cloud compute power was deemed to be the best option and over 5 years the TCO would be under 25% rather than buying in hardware. The environment has been implemented in such a way that we can scale up or scale down if there is a requirement in the future.
![How To Thow A Wiffle Ball Curve [Step-By-Step]](https://i.skyrock.net/5231/101335231/pics/3302120606_1_2_oUUQF1lp.gif)
