GIS x machine for big data processing

A customer involved in public health research asked us for a machine for processing GIS and big data.
I would like a workstation that can perform MCMC processing, natural language processing with Python (PyTorch), network analysis with a geographic information system, etc.
In terms of specs, we believe that a high-precision GPU and a CPU that can handle big data processing are necessary, and we specifically envision the following.

・Memory: About 128GB at the time of shipment.Configuration that can be expanded up to 1TB.
・GPU: Equipped with 6000 NVIDIA RTX A2s and equipped with NVLINK
・Software used: R, Rstudio, Python, ArcGIS, QGIS
・Others: Quiet enough to withstand office use

Based on the content of the consultation, we proposed the following configuration.

CPU Xeon W7-3455 (2.50GHz 24-core)
memory 128GB
Storage 1 4TB M.2 SSD
Storage 2 16TB HDD S-ATA
video NVIDIA RTX A6000 x2
network on board (1GbE x1 /10GbE x1)
Housing + power supply Tower type housing + 1600W
OS Ubuntu 22.04
Others NVLINK Bridge

It is configured with a 3400-core model of the Xeon W-24 series.
Equipped with two NVIDIA RTX A6000 and includes NVLink Bridge.

It is assumed that the memory will be expanded from the initial 128GB to 1TB in the future, but all memory with 128GB must be removed.Therefore, the memory module configuration at the time of initial shipment does not consider memory expansion and prioritizes using all memory channels.

Storage is provisional capacity.
For use in big data processing, if you let us know the approximate storage capacity required for the machine itself, we will change it accordingly.
Also, since ArcGIS uses storage for the cache area, the SSD on which the OS and software are installed is a faster NVMe compatible product.

Considering that the installation location is in the office, we have adopted a highly silent housing, but when the video card is fully operated with GPGPU, a certain amount of driving noise will be generated.If you want even more cooling performance and quietness, we also have an optional silent rack, so please contact us.

For customers who want measures against operating noise,SI companyWe accept proposals in combination with our silent racks.
Please feel free to contact us if you are considering installing a machine that emphasizes quietness.

Features of silent racks manufactured by SI
[1]Provide a specially designed rack that matches the user's environment and machine
[2] High-level compatibility between quietness and safe heat dissipation
[3] Because it is a manufacturer centered on specialized acoustic technology, it has high quietness technology
[4]Promise safe operation with technical service for machine adaptation

In addition, the configuration of this example is a configuration that prioritizes performance regardless of the amount of money.
Since the number of memory channels is 8 and it is high performance, it is thought that the first point to review for cost reduction is around the CPU and memory.Please feel free to contact us for a plan proposal that fits your budget.

 

■ Keywords

・What is R?
R is an open source and free software programming language/development execution environment for statistical analysis.Used in calculations and graphing for statistical processing.
Since many libraries exist, complex techniques can be handled by simply calling the library.

reference:The R Project for Statistical Computing *Jumps to an external site

 

・What is Rstudio?
Rstudio is an integrated development environment for using R.It offers an intuitive and easy-to-use user interface, including project management features, code editor, code auto-completion, syntax highlighting, debugger, profiler, support for markdown documents, package management features, interactive graphics viewing, and more. It has a variety of functions.

reference:RSTUDIO IDE (Posit) *Jumps to an external site

 

・What is Python?
Python is an object-oriented programming language copyrighted by the Python Software Foundation (PSF).Its programming syntax is simple, making it highly readable, and it also features a wide variety of components, such as libraries and frameworks, that are suitable for different purposes.A popular language for programming beginners to advanced users.

reference:Python *Jumps to an external site

reference:[Feature Article] Programming language Python Why is it so popular? --Tools to accelerate Python programming * Jump to our owned media "TEGAKARI"

 

・What is ArcGIS?
ArcGIS is the leading software platform for geographic information systems (GIS).It can collect, manage, analyze, and visualize geographic data, and is used as GIS by many organizations around the world.ArcGIS Pro and ArcMap as desktop products, ArcGIS Enterprise as server products, AppStudio and ArcGIS API for Python for developers, and ArcGIS QuickCapture and Explorer for mobile.It is widely used as GIS by administrative agencies and companies around the world.

reference:What is ArcGIS? (ESRI) *Jumps to an external site

 

・What is QGIS?
QGIS is an open source GIS software developed by the QGIS Development Team. It supports visualization, editing, and analysis of spatial data with an interface similar to ArcGIS.Its functions are limited compared to ArcGIS, but it is suitable for personal and educational purposes due to its free use and community support, and is used worldwide.

reference:QGIS (ESRI) *Jumps to an external site

 

・What is NVLINK?
NVLink is a high-speed interconnect technology between GPUs developed by NVIDIA. GPUs with NVLink achieve faster communication bandwidth and speed than PCI Express, greatly improving the efficiency of data transmission and reception between GPUs. NVLink is sometimes used as a technology to improve the efficiency of large-scale parallel processing in the fields of AI and high-performance computing.

reference:NVIDIA NVLink (NVIDIA I) *Jumps to an external site