A customer involved in marketing research consulted us about a machine for statistical analysis.
We would like a machine to perform the MCMC method in Bayesian statistics using R (RStan), focusing on statistical analysis of consumer questionnaire data.The maximum number of questionnaires that can be handled is about 1,000.
In addition, as a method, Bayesian network analysis using BayoLinkS and Bayesian network analysis using R are assumed, and the specific conditions are as follows.
・CPU: Core i9
・ Memory: 64GB
・Storage: SSD 1TB
・OS: Windows 11
・Display: 23.8-inch liquid crystal display
・Software used: R, Rstudio, Python, BayoLinkS
・Others: I want to be able to comfortably execute four calculations by the MCMC method in parallel.
・Budget: About 50 million yen
Based on the contents of the communication, we proposed the following configuration.
|CPU||Core i9-13900KS (3.20GHz 8 cores + 2.40GHz 16 cores)|
|storage||2TB M.2 SSD|
|video||on board (DPx1 HDMIx1)|
|network||on board (2.5GBase-T x1) Wi-Fi x1|
|Housing + power supply||Middle tower case + 850W|
|OS||Microsoft Windows 11 Professional 64bit|
23.8-inch wide FullHD LCD display
Based on the use of R and the budget conditions, I think that the Core i9 CPU is good as you requested.
4 If it is used for calculation, there is a high possibility that the clock boost will be effective, so even the 13th generation Core i9 has selected a product with high single core performance. In the results of running only P cores, we confirmed operation at around 8 GHz with 5.6 cores running.
Compared to the 12th generation Core i9-12900K, there is a difference of about 33% (12% for single) as a CPUMark value.
There is an option of ECC memory, but considering the usage, it is hard to say that 64GB is essential, and it will lead to an increase in cost, so we chose NonECC specification. Since the CPU itself supports ECC memory, it can be changed if the cost increase is acceptable.If you expect continuous calculation at the level of several weeks, we recommend using ECC memory.
Also, if there is a possibility that R-type intermediate files may be exported to storage, a larger SLC area controlled by the SSD side as a cache will lead to improved performance, so we chose a product with a capacity of 2TB. It is a high-speed type compatible with PCI-E Gen4.
The configuration of this case study is based on the conditions given by the customer.
Please feel free to contact us even if you are considering different conditions from what is posted.
・What is R?
reference:The R Project for Statistical Computing *Jumps to an external site
・What is Rstudio?
reference:RSTUDIO IDE (Posit) *Jumps to an external site
・What is Python?
・What is BayoLinkS?
reference:BayoLinkS (NTT DATA Mathematical Systems) *Jumps to an external site
・What is Bayesian network analysis?
■ Click here for details and inquiries about this PC case
* Please enter the name of the case or your desired conditions.