As we saw in the previous two blogs, Amazon’s AWS can have sudden and unexpected problems. Some of these can’t be detected with existing monitoring solutions. Users need a light weight monitor, with a small cpu and memory footprint. Users decry complexity with lots of bells and whistles, instead to simply know 3 things:
1) How good is the cloud server that AWS is offering me?
2) What is its performance or how much time my application will take to run on it?
3) How soon will I know if the server performance changes?
We embarked to find answers and came up with Meghafind. “Megha” is a Sanskrit word for Clouds, and to find the right cloud machine for you is what we do, hence Megha+find = Meghafind.
Our solution comprises of Scouts, which are small programs running at a user-level in a VM, or on any operating system, and on any machine of your choice. Scouts are able to accurately determine a machine’s performance items of its CPU, Memory and Storage components. The resulting scout score on a server can be compared with another server’s score, or for the same server with a different hardware configuration, or with different loads at different times.
Our solution collects scout results for comparisons across systems and configurations. We have unique data-mining and state of the art machine learning algorithms to predict performance of your application on a new server, just by running the scouts on it. Small footprint of Meghafind scouts ensures that your server performance is not altered by our measurements. Furthermore, we provide multi-threading support to determine optimal configuration for your application to extract the best performance on any given server.
We welcome you to try, by visiting the install page and play with scouts hands-on, or send us a note in the Contact form. An account and password is required for saving and comparing your scout results, that we will issue after hearing from you. Happy Scouting.