I have this project about atherosclerosis and huge amount of computation. I don't know how can I use parallel computing in ADINA or using a cluster or distributed computation. The information in the internet is so confusing. Anyone know an introductory book?
Hi Morteza,
If you're building an application from scratch, the first thing you should consider when deciding what architecture computing platform you should use is the algorithm that the application implements. This will help you to decide what level of parallelism you should aim the implementation of the application at. If the application is highly parallel, but tightly bound between computational domains, you should go for a cluster-like solution. If the algorithm is something more like massive monte-carlo, and computational domains are loosely coupled or not coupled at all, you would do well do use a massively distributed architecture.
It really depends on what you have available to you though.
Checking on the web about ADINA, it looks like you are simulating a tightly-bound system, which will probably be efficient in a cluster environment.
If I were you, I would think about the full workflow of the problem, from start to finish. What is the basic research question ? In order to arrive at the answer, will I have to do a parameter sweep (which is massively distributed) or is the analysis more linear ? If you can write down this in plain English (or some other human language), it will be far easier to choose the right tools for the job.
Paralization in ADINA is only for equation solver which is in most cases not the main reason for high demand capacity
ADINA have options to run either as SMP or on a cluster as DMP. But there is a separate code and licensing for DMP.
- Badges
- Science method