TEGAKARI
  • Home
  • Overseas Products What's New (Unipos)
  • R & D PC configuration example (Tegsys)
  • Service information for R & D
    • Rental service tegakari
  • Technical information articles
  • Version upgrade information
  • News from TEGARA
  • Contact
Pickup new articles
  • [April 2025, 5] Next Generation Sequencing (NGS) Data Analysis Workstation Research workstation
  • [April 2025, 4] Bioinformatics Workstations Research workstation
  • [April 2025, 4] Machine for the crystal structure analysis software suite "CCP4" (April 2025 version) Research workstation
  • [April 2025, 4] GAMESS(US) workstation Research workstation
  • [April 2025, 3] Special offer! Post-purchase support included: New fiscal year bio-related software campaign Hot topics now

Home > Humanities / Social Sciences > [PC knowledge] How to choose?PC for data science [Stata edition]

[PC knowledge] How to choose?PC for data science [Stata edition]

2023/ 2/ 13 TEGARA Co., Ltd. Research workstation, Humanities / Social Sciences, Medicine / Nursing / Pharmacy, Biology / Agriculture, Informatics, Overseas Products What's New (Unipos), R & D PC configuration example (Tegsys)

 

[Please check] This is the following articleSequel articleWill be

[Product introduction] Tools useful for data science [Stata edition]

In research and development, you may use specific software.In fact, the technical requirements of each software are surprisingly rough.Also, the software used is not limited to one.In other words, it can be difficult to match or optimize the performance of multiple software and machines based on a single requirement or request.

So this time,Last articleLet's take Stata, which is useful for data science introduced in , as an example, and introduce "points for selecting a PC".

Custom-made PC production service "Tegusis” and overseas product procurement and consultation services for research and development “UniposIt is possible only because Tegara provides two types of services, ""Purchasing overseas products for R & D and manufacturing custom PCs optimized for their useis an example of a one-stop service.

table of contents

  • Need to have the right machine?
    • For example, such consultation
    • And in other words, what are the specs required for the PC?
  • What are Stata's system requirements?
  • PC selection points using Stata/MP as an example
    • About Stata / MP
    • How can I review my PC?
  • Summary: PC design points for Stata/MP
  • Next notice

 

Need to have the right machine?

When you ask us about a PC for data science, machine learning, or deep learning, do you ever worry about what kind of PC you should have?
For "what you want to achieve", the PC is just a "tool" that assists it.The selection of the most suitable tools is essential to achieve the goal and is the shortest route.

For example, such consultation

  • I use Stata, but I feel that the processing speed is slow
  • I bought a high spec machine, but it doesn't meet Stata's requirements
  • Stata is working fine, but other processes take a long time
  • I want to use software other than Stata

 

And in other words, what are the specs required for the PC?

Here are the "top 3 requirements" that you request not only for problem solving, but also for what you can do.

  • processing power
  • Cost performance (cost effectiveness)
  • Achieve stability and scalability

At our company, we consider the configuration with an emphasis on the most prioritized requirements within the budget.
Then, in order to realize these requests as much as possible, I would like to derive the configuration from both software and hardware.

What are Stata's system requirements?

In recent years, CPUs have become many-core, and general CPUs support parallel processing (a method of shortening calculation time by processing multiple calculations at the same time).If you want to improve the processing speed by parallelization, configure the hardware with emphasis on the number of CPU cores.

Stata has three editions: Stata/MP, Stata/SE, and Stata/BE, depending on the data set (aggregation collected for a certain purpose or subject).Stata/SE and Stata/BE, which do not support parallel processing, adopt a CPU that emphasizes the clock.

Stat/MP, on the other hand, is programmed to take full advantage of multi-core and multi-processor computers, so it emphasizes the number of physical cores in the CPU.

First, check if the command you plan to use supports parallel processing.

  • Extensive use of serial processing : Stata/SE or Stata/BE, and Stata / MP
  • Heavy use of parallelism : Stata / MP

Prioritize the command to process.We will also consider processing speed.

  • Don't worry about processing speed : Select Stata/SE or Stata/BE (clock sensitive)
  • Fast processing speed is important : Select Stata/MP (emphasis on number of cores)

 

In addition, when selecting each edition, it is necessary to select Stata according to the number of variables to be handled.Please check the following table.

Edition variable maximum independent variable maximum Maximum value of observed data Parallel processing Number of cores RAM
Stata / MP 120,000 65,532 200 billion * 〇 2 cores or more 4 MB
Stata / SE 32,767 10,998 21.4 billion × 1 core 2 MB
Stata / BE 2,048 798 21.4 billion × 1 core 1 MB

 * The maximum value of observed data is limited by the amount of RAM available on your system.It is recommended that the amount of RAM be 1.5 times the amount of data to be used.

PC selection points using Stata/MP as an example

From here, we will introduce how to choose a PC, assuming that Stata/MP is selected with emphasis on the speed of processing speed.

About Stata / MP

For Stata/MP, the license to be selected changes according to the number of cores used.For example, if your machine has 8 physical cores, choose an 8-core, 4-core, or 2-core Stata/MP license, depending on how many cores you actually use.Consider the number of cores for your machine and license depending on how important processing speed is.

Stata/MP distributes many of Stata's most computationally intensive tasks across all cores of your computer, making it faster than any edition.In general, if you want to increase the processing speed, you need to be careful when selecting parts such as the CPU, memory, storage, and video card (*), but in the case of Stata/MP, the core of the CPU is important.Other hardware requirements are not very high.

* CPUs with a large number of cores tend to have a low clock speed.As a result of prioritizing the number of cores too much, the processing speed may become slower.For this reason, it is necessary to consider the balance, such as adopting the CPU with the highest clock speed from among the CPUs with the number of cores close to your desired number.

See Stata's system requirements here

Compatible operating systems
https://www.stata.com/products/compatible-operating-systems/

 

Below are articles related to Stata

TEGAKARI
2021.09.16
[Special feature article] How-to on how to select a "computer" for research and development that supports science and technology
https://www.tegakari.net/2021/09/sp-how-to-choose-the-pc-for-scientific-computing
Strengthening science and technology capabilities is considered to be the most important issue for Japan to survive in the world in the future, and the recently announced increase in the budget request for science and technology-related budgets for the next fiscal year (FY4)...

Stata/MP is the fastest and largest edition of Stata.
https://www.stata.com/statamp/

 

How can I review my PC?

First, open Task Manager and check the number of CPU cores.

Start and check task manager

For Windows 11, press "Ctrl + Alt + Delete" at the same time and select "Task Manager" from the window menu.If you select the Performance tab in Task Manager, you can see the number of cores in the lower right part.
For example, if your machine has "4 cores", it is desirable to select a 4 core license.

 

Why match the license with the number of cores?

performance stata region

Stata/MP requires about 2-33% for analysis on a relatively inexpensive 50-core PC and about 4-50% for a standard 75-core computer.You can save time. For a 6-core license, if the number of physical cores of the CPU is less than the number of cores originally available for computation, the computing power will be reduced accordingly.Therefore, if you have a 6-core license, you will need a machine with a CPU with 6 or more cores to fully utilize the license.

Number of cores 2 core 4 core 8 core
all commands x1.7 x2.4 x3.1
Estimated command (median) x1.8 x2.9 x4.1
Logistic regression x2.0 x3.8 x6.9
finite mixture model x2.3 x3.5 x4.5
Normality test (90%) x3.6 x4.4 x5.8

"Performance per command for Stata/MP" is published in the following manufacturer report (page 17 onwards).Please refer to

Stata/MPPerformanceReport
https://www.stata.com/statamp/report.pdf

 

The more licenses/cores, the faster the process

  • I want to handle even larger datasets
  • I want to speed up processing

If there is a request like the above, the more cores the CPU has, the faster the processing can be achieved.

We can propose a set of licenses and machines according to your budget.We will select a license after considering your budget and propose a machine configuration.Please feel free to contact us.

■ Click here for details and inquiries regarding Stata machines
Machine for Stata

 

Is it OK if the number of CPU cores exceeds the license? !

For example, if you compare machines with 32 or 32 core CPUs for a 64 core license, the machines with 32 core CPUs may be faster.

In Stata/MP, there is a characteristic that the processing speed decreases as the number of licensed cores and the number of physical cores increases.For this reason, it cannot simply be said that the more physical cores a CPU has, the better.

If you purchase a 64-core license in the near future, a machine with a 64-core CPU is also an option, but the machine with a 32-core CPU is the best choice for a 32-core license. Recommended.

If you do not want to execute parallel processing, select a CPU based on the clock rather than the number of cores.

 

Summary: PC design points for Stata/MP

PC design points for Stata/MP

  • is a multi-core and multi-processor computer
  • The number of physical cores of the CPU is 2 to 64 cores
  • The number of license cores is about the same as the number of physical cores of the CPU
  • Large amount of data to process
  • Use many commands for parallel execution (*)

*Please check if the command you plan to use supports parallel processing.
Stata features are common across all three editions (Stata/MP, Stata/SE, Stata/BE).If you use a lot of serial execution commands, it may be better to choose Stata/SE or Stata/BE.

 

Next notice

Next time, we plan to release [Proposal example] PC for data science [Stata edition].

■ Click here for details and inquiries about Stata
Stata | Statistical Software Package
Manufacturer (StataCorp LLC) WEB site

 

  • stata
  • Data science
  • Statistical analysis
  • Non-linear analysis
  • analysis
  • Parallel computing

People who read this article also read this article

Mathematical Science

[Product introduction] Boulder Opal and Black Opal | Q-CTRL quantum control platform

2024/ 1/ 31 TEGARA Co., Ltd. Mathematical Science, Informatics, Business support and efficiency tools, Overseas Products What's New (Unipos)

A page for Q-CTRL's next-generation quantum computing products "Boulder Opal and Black Opal" has been added to the Unipos website. Q-CT […see next]

ideal computer for stata
Humanities / Social Sciences

[Configuration example] PC selection flow and its points [Stata edition]

2023/ 3/ 28 TEGARA Co., Ltd. Research workstation, Humanities / Social Sciences, Medicine / Nursing / Pharmacy, Biology / Agriculture, Informatics, Overseas Products What's New (Unipos), R & D PC configuration example (Tegsys)

[Please check] This is a sequel to the article below [Product introduction] Tools useful for data science [Stata edition] [PC trivia] How to choose?for data science […see next]

Humanities / Social Sciences

RStudio, an integrated development environment (IDE) for the statistical programming language R

2019/ 11/ 8 TEGARA Co., Ltd. Humanities / Social Sciences, Informatics, Application development and programming, Overseas Products What's New (Unipos)

■This article was posted on November 2019, 11, so the information may be out of date.An integrated development environment for the statistical programming language R on the Unipos website […see next]

Site search:

Tegara YouTube Video

[Effect of IR Pass Filter] Shoot whiteboard with RealSense D435 and D435f

The latest posted video is displayed.
Other videosTegara Corporation Youtube channelplease look at

Popular Articles (Access ranking for the last 7 days)

  • VisualSVN Server [Product Introduction] VisualSVN Server | Subversion server software for Windows 2023/ 10/ 23
  • [Release information] Remote access RealVNC VNC Connect | Notice of license change 2023/ 6/ 29
  • [Product introduction] Leap Motion Controller 2 – Hand tracking camera that recognizes hand and finger movements 2023/ 6/ 9
  • Illustration tool "BioRender" for the life science field 2021/ 9/ 30
  • furix BetterWMF and CompareDWG tools for AutoCAD [Product introduction] Beyond Compare: File and folder comparison, integration and synchronization utility 2022/ 11/ 18

Latest posts

  • TEGSYS Next Generation Sequencing (NGS) Data Analysis Workstation
    Next Generation Sequencing (NGS) Data Analysis Workstation
    2025/ 5/ 15
  • Bioinformatics Workstations
    2025/ 4/ 22
  • Machine for the crystal structure analysis software suite "CCP4" (April 2025 version)
    2025/ 4/ 22
  • GAMESS(US) workstation
    2025/ 4/ 22
  • Special offer! Post-purchase support included: New fiscal year bio-related software campaign
    2025/ 3/ 28

Featured tags

Analysis tool (56) 3D camera (55) Machine learning (machine learning) (53) AI (47) Robotics (45) VR (44) Robot arm (42) RealSense (41) Statistical analysis (39) Bioinformatics (39) Video / Video (37) SBC (36) Depth camera (36) Deepearning (36) instrumentation (35) IoT (35) Small SBC (35) simulation (33) Spectrum (33) Data analysis (31) Python (29) Cyber ​​security (28) Next-generation sequencer (27) JavaScript (27) AR (27) Chemical (27) First principle (26) . NET (26) In-vehicle (25) TO DEAL (25) Image processing (25) Metashape (25) MATLAB (24) UI (24) Photogrammetry (23) Educational robot (22) 3D model (22) Image analysis / image inspection (22) prototype (22) Molecular biology (22) Support (22) Web development / production (21) Measuring instrument (21) Test tool (20) material (20) GIS (20) security (19) Psychology (19) Mech robot (19) Animation (19) Mobile robot (19) ROS (19) Robot hand (19) Visualization (19) Drone (19) protocol (18) programming (18) ToF (18) Autonomous vehicle (18) Electromagnetic field analysis (18) EEG (18) CAE (17) Clinical (17) Motion capture (17) gene (17) 3D printer (17) Raspberry Pi (17) tracking (17) Deep learning (17) DNA (16) XNUM XD modeling (16) Structural analysis (16) chart (16) Industrial (16) Education (16) Bioassay (16) modeling (16) Movie editing (15) Arduino (15) 3D scan (15) AR / VR (15) biostatistics (15) Library (15) drug development (15) Molecular dynamics (15) Fluid analysis (15) 写真 (14) Information dissemination September issue (14) Stimulus presentation (14) CFD (14) Device control (14) others (14) SLAM (14) Malware (14) Articles delivered in August 2022 (14) CUDA (14) Articles delivered in August 2022 (14) control (13) Numerical analysis (13) Thermal fluid analysis (13) Monitoring (13) 24 hours operation (13) STEM / STEAM education (13) Development and evaluation kit (13) 3D CAD (13) Voice processing (13) Surveying (13) Depth sensor (13) wireless (13) Nanostructured material (13) Agriculture / Agriculture (13) IDE (Integrated Development Environment) (13) Information dissemination February 22 issue (12) natural Science (12) Information dissemination February 22 issue (12) Remote operation (remote control) (12) Capture glove (12) Looking Glass (12) Genome analysis (12) CAD (12) FDTD method (12) GPGPU (12)
Find Information by Field-Category
  •  Humanities / Social Sciences
  •  Mathematical Science
  •  Chemical
  •  engineering
  •  Medicine / Nursing / Pharmacy
  •  Biology / Agriculture
  •  Informatics
 
  •  Artificial intelligence
  •  Robotics
  •  Sensor technology
  •  Development kit / electronic work
  •  Digital gadget
  •  Automotive / vehicle related
  •  Industrial communication technology
  •  Application development and programming
  •  Network security
  •  Multimedia (video / image / audio) processing
  •  Business support and efficiency tools
Translate
English English Japanese Japanese
Contact Form – Contact
Click here to contact TEGAKARI
Site link
Privacy Policy
Management website (service)
TEGARA Co., Ltd.
TEGARA CORPORATION corporate site

UNIPOS
Overseas product procurement and consultation services for R & D

Tegusis
Research and industrial PC production and sales services
SNS account
  • Twitter
  • YouTube
  • Facebook

TEGARA Co., Ltd.

Tegara is a platform that provides R & D with useful products, services, and information in an integrated manner. "Helping accelerate R & D"

Copyright © 2020 | Tegara Corporation