When everyone digs for gold, sell shovels.
The rapidly increasing prices of NVIDIA stocks are the talk of the town.Nvidia’s value has skyrocketed, hitting $3 trillion in June 2024 and surpassing Apple to become the second most valuable U.S. company. Just last year, Nvidia was worth $1 trillion and was already more valuable than Amazon and Google’s parent company, Alphabet.
Let’s look at what NVIDIA is into?
Nvidia dominates the market for Graphic Processing Units (GPU). For a long time NVIDIA, primarily sold its GPUs to professional graphic artists and video gamers. Later they had found new markets for chip technology, namely artificial intelligence, data science, crypto currency mining, high performance computing etc. All these had fuelled incredible growth for business. Nvidia shares have soared because of the accelerating demand for its graphics processors.
These chips are essential to the training of complex AI models. NVIDIA began focusing significantly on artificial intelligence around 2012.
This shift was marked by their work with the CUDA (Compute Unified Device Architecture) platform, which allowed developers to use the parallel processing power of Nvidia GPUs for general-purpose computing tasks and AI.
In 2012 they have released Kepler GPU architecture which significantly improved compute performance. But the real milestone happened with the introduction of Pascal architecture specially designed to enhance AI & Deep learning workloads.
They will set a giant leap with release of latest GPU architecture ‘Blackwell' (2024).
NVIDIA CEO Jensen Huang’s latest keynote at the GPU Technology Conference (GTC), witnessed several major announcements.
‘Blackwell' architecture is described as the “World’s most powerful chip”. Claiming it is designed for large scale AI and generative AI applications.
Jensen Huang introduced NVIDIA Inference Microservices (NIM), a new method for developing AI software. In which developers use AI models that learn from examples and feedback to generate codes. This approach leverages NVIDIA’s computing libraries and generative AI models to simplify and enhance software development .
Their Project GR00T is focused on advancing AI and robotics. It includes the Jetson Thor computer, which is made for AI-powered robots, and the Isaac robotics platform, which improves how robots see and handle objects.
NVIDIA showcased a 3D blueprint for next-generation AI data centers, emphasizing the importance of these facilities as "AI factories" that generate intelligence.
Tech Giants Dependancy
NVIDIA started to concentrate heavily on AI and deep learning around 2012. But their major advancements and dedicated AI hardware emerged from 2016 onwards.
Tech giants like Google, Amazon, and Microsoft heavily rely on NVIDIA for their AI and cloud computing needs.
Google uses NVIDIA GPUs to speed up its AI and machine learning tasks. These GPUs are crucial for Google’s AI systems, including its TensorFlow framework and other research projects. This partnership helps Google offer advanced AI services and boosts the performance and efficiency of its data centers.
Amazon Web Services leverages NVIDIA GPUs to drive its Elastic Compute Cloud (EC2) instances, which are vital for customers requiring high-performance computing for AI, machine learning, and data analytics.
Microsoft Azure also rely on these GPUs to support machine learning services and contributes to development of AI models.
In short NVIDIA is providing essential tools and services that supports the current ‘AI wave’.
Hope now you have understood the context “selling shovels during a gold rush”
Happy reading folks.
