Data Science Shop’s Post

The adoption stage of the #DataProductCycle focuses on identifying potential obstacles to adoption and creating strategies to overcome them. Read more in our #Substack: https://lnkd.in/ema7kBCu Marco A Morales, PhD #DataScienceShop #dataproducts #dataengineering #mlengineering

Once the problem is clearly defined and data availability confirmed, we move into the heart of the #DataProductCycle: the development and implementation stage. Read more in our #Substack: https://lnkd.in/ema7kBCu Marco A Morales, PhD #DataScienceShop #dataproducts #dataengineering #mlengineering

Horizontal scaling is a powerful technique but always keep in mind Amdahl’s law which states that the overall speedup of a system is limited by the portion that cannot be parallelized or improved. This has direct implications for horizontal scaling, where additional compute resources (nodes/servers) are added to a system. For example, let's say 80% of your application can be parallelized across multiple nodes, while the remaining 20% is sequential. Even if you horizontally scale by adding an infinite number of nodes, the maximum theoretical speedup would be 5x (1 / (1 - 0.8) = 5). This is because the sequential 20% portion acts as a bottleneck, capping the overall speedup. This law also applies to project management where blindly throwing more people at a project may not yield linear gains, especially if a significant portion is inherently sequential work like integration, testing, management etc. Amdahl's law highlights the importance of identifying and optimizing these sequential bottlenecks before scaling the parallelizable work across more resources. #compute #organizations #formula #architecture

In this informative article, Bentley Chief Technology Officer Julien Moutte shares how the combined capabilities of iTwin and Cesium’s open-source offerings mean even more innovation and collaboration in the AEC industry. #bentleysystems #innovation #cesium #geospatial

QSLS introduces a new dimension of engineering…measuring levels of support for Characteristics, Quality Attributes and Business Drivers for a system starting with Architecture.

👇 Learn how XYZ Reality is changing the game in construction 👇

System of systems is an emerging field where dedicated systems combine to create a more complex system with superior functionality and performance. However, the methodology for defining, modeling, and analyzing system of systems problems remains incomplete. One critical missing aspect is the lack of inclusion of the environment into the definition of the system, as "System in an Environment" (SiE). Keeping a SiE secure is just a matter of placing it in a safe environment. Learn more about system of systems engineering on Wikipedia. #systemofsystems #engineering #environment #research

Do you use foreign key relationships in your production databases? If so, what benefits do you see from using them? If not, what are the reasons for avoiding them? Curious to hear from the community! #engineering #techdiscussions

 Day 210 🌟! Exciting updates: Made strides in enhancing the search bar of my IMDb clone project. Delving deeper into Engineering 🚀, embracing #TechTrends and #Innovation. Committed to professional growth with #ProfessionalDevelopment. Stay tuned for more! #IndustryNews #EngineeringLife #Design 💻 #FutureOfEngineering #RenewableEnergy #Automation 🔄

The Diagnosis stage is the foundation of the entire #DataProductCycle and doesn't receive enough attention. It's about identifying a problem, defining it with crystal clarity and assessing whether a data-driven solution is feasible. Read more in our #Substack: https://lnkd.in/ema7kBCu Marco A Morales, PhD #Datascience #DataScienceShop #dataengineering #mlengineering