The Exaflop Super-computing Scale of Cudos | by Cryptoruppted | Coinmonks

In order to provide quicker innovation, adaptable resources, and scale economies, cloud computing is the distribution of computer services over the Internet (“the cloud”), including servers, storage, databases, networking, software, analytics, and intelligence. With its emergence, big firms can access anything from programs to storage from a cloud service provider, renting access to them rather than owning their own computing equipment or data centers. And, conversely, Users can avoid the upfront cost and complexity of purchasing and maintaining their own IT infrastructure by using cloud computing services instead, and they only pay for what they use, when they use it.

Delivering the same services to a large customer base allows cloud computing service providers to realize enormous economies of scale. The need for a more sustainable and decentralized source of computing power is evident when we consider the AI-driven society in which we already live, where organizations like ChatGPT, OpenAI and others are continually training models. More data is being crunched by computers every second than ever before in an effort to solve the most challenging issues of our time, including how to treat diseases like COVID and cancer, limit climate change, and more.

This is just where Cudos comes into play, especially in decentralised cloud computing.

Cudos is a Delegated Proof-of-Stake, layer-1 blockchain with a vision for a decentralised computation accessible on-chain. To enable safe, decentralized, and permission-free access to high-performance computing at scale, the Cudos network’s design separates consensus from execution. With the goal of building a decentralized, sustainable, and connected world, Cudos combines cloud and blockchain technology. These and other major problems propelled computing into the exascale era we are in today, where top performance is frequently expressed in exaflops.

What exactly Is an Exaflop, then?

A supercomputer that can do at least 10 to the power of 18, or one quintillion, floating point operations per second, is said to have an exaflop. This term is used to describe how well supercomputers and other high-performance computing systems function. The exa- prefix in exaflop stands for a quintillion, also known as one followed by 18 zeros or a billion billion. Like a quintillion bytes of data, an exabyte is a memory subsystem. Floating point operations are indicated by the “flop” in exaflop. Exaflop/s represents the number of flops that a system can execute in a second.

The number of floating-point or flop operations per second is a crucial indicator of computational efficiency. It is the foundation for the yearly Top 500 supercomputer rankings and applies to calculations where all numbers are expressed with decimal points.

Exaflops are equal to x1⁰¹⁸ flops. That translates to “a billion billion calculations” per second, according to the New York Times. Exaflops are equal to 1,000 Petaflops.

The prefix peta- denotes x1⁰¹⁵, or a number with 15 zeros after it. So, a petaflop is one exaflop.

According to research, an exaflop system would use between 150 and 500 MegaWatts of power. Comparatively, 500 MW is about similar to the amount of energy utilized by a small city with a population of about 500,000, to put it in context.

Builders on the Cudos network get access to:

🔧 High application performance

🏗 Developer-friendly toolkit

💱 Composable asset management

Users of the chain benefit from:

🔐 High cryptographic security

🔬 Public blockchain transparency

📬 Cross-chain interoperability

🌳 Carbon-neutral transactions

Proof-of-Stake Consensus

A Tendermint Core engine manages consensus — powered by a Byzantine Fault Tolerant (BFT) consensus model. The BFT model can tolerate less than one-third of the Byzantine validators reaching consensus. Considered to be a gold standard in Proof-of-Stake consensus, it is the most widely used consensus engine across the industry.

Twitter | Website
Discord | Telegram
YouTube | LinkedIn
Spotify | Medium
Facebook