Analysis machine specialized for Stata/MP operation

Case No.PC-9133BA customer who viewed this requested a machine for analyzing tax data with Stata/MP.
The data file of Stata/MP is about 170GB, and we would like to introduce a machine that can handle this large size file without problems with a budget of about 300 million yen.
More specifically, even if the number of variables handled by Stata/MP increases by about three times during program execution, and the required memory increases accordingly, we have received a condition that the specifications can handle without processing dropouts.

The machine specs that the customer is currently using are as follows.

・CPU: Core i5-8400 (2.8GHz 6 cores)
・ Memory: 64GB
・OS: Windows 10

You asked for 512GB of memory, but since data files of nearly 20GB are added every year, we would also like to have a proposal to ensure memory expandability.
Customers have also asked us, "I think it's better to use SSD for storage considering the memory overrun when dealing with Stata/MP, but if there's no particular need to consider it, please let me know." increase.

Based on the above conditions, we proposed the following configuration.

【Main Specifications】

CPU AMD Ryzen ThreadripperPRO 5975WX (3.60GHz 32 cores)
memory 512GB REG ECC (64GB x8)
Storage 1 4TB M.2 SSD
Storage 2 16TB HDD S-ATA
video Nvidia T400
network on board (1GbE x1 10GBase-T x1)
Housing + power supply Tower type housing + 1000W
OS Microsoft Windows 11 Pro 64 bit

■ Points

・The CPU selection for the Stata machine depends on the number of cores contracted for the Stata license.
・Considering the speed between the CPU and memory, it is recommended to install in a way that fills all the slots without creating any empty space in the memory slots.

Since the customer inquired after looking at the PC-9133B (32-core license specification configuration), we considered it assuming a 32-core contract.

The memory is 64 GB with 8 x 512 GB modules that do not create empty slots, giving priority to speed.
The maximum memory capacity is 1TB, so if you want future expandability,PC-10746B.
If you want a configuration with 1TB of memory,PC-10746C.

For storage, the system side is a high-speed SSD in case the amount of data to be handled overflows from the memory.
However, due to the nature of SSD, there is a limit to the number of times it can be rewritten, so it is not recommended to use it on the premise of constant memory swapping.
Since speed and durability are affected by the amount of free space on the SSD, we select products with large capacity.However, instead of leaving the main data on this SSD, it is assumed that it will be used by moving it to the 16TB HDD side and saving it.
On the SSD side, enough free space is left to ensure sustain performance, so it is assumed that unnecessary data is not placed on the C drive as much as possible.

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.

 

■FAQ

・What is Stata
Stata is a comprehensive statistical software package with features such as data analysis, data management and chart generation.
In addition to GUI operation with a mouse, it can also be executed with a powerful and intuitive command syntax, making it easy to use, fast and accurate.

reference:Been * Jump to our Unipos handling product page

 

・What is memory swap?
The act of moving data in memory to storage when memory space is scarce.
When the data moved to the storage is needed, it is necessary to recall it to the memory again.
This operation itself is automatically performed on the OS side, but if swapping occurs frequently, the operation of the PC may become unstable.
Also, since storage reads and writes occur frequently, using the SSD as a swap area can shorten the write life.

 

・What is sustain performance?
The performance of continuous data transfer in a storage device that uses an HDD or SCSI bus.