#internet scale

Elrond is a low-cost, high-performance, and scalable blockchain network that can be used by both dApp developers and enterprise solutions.

Summary

Elrond is an internet-scale blockchain that is designed to be fast, cheap, and efficient across a wide range of use cases. Elrond Network offers industry-leading scalability because of Adaptive State Sharding and its unique Secure Proof-of-Stake (SPoS) consensus method.

The sophisticated aspects of the Elrond crypto protocol are highlighted in this article.

Contents

Elrond Network Design

The Elrond Coin: eGold (eGLD)

Secure Proof of Stake (SPoS)

Elrond’s Blockchain Sharding Innovations

Smart Contracts on Elrond Crypto Network

Elrond Use Cases

>>The Best Bitcoin and Cryptocurrency Exchange Guide<<

Elrond Network Design

The Elrond blockchain, which will be launched in July 2020, is meant to provide distributed apps and organizations with decentralization, security, and scalability.

The project aims to achieve maximum operational efficiency by focusing on cross-chain interoperability, developer assistance, and cross-chain interoperability.

Elrond Network refers to itself as an internet-scale blockchain, implying that it is built to manage transactional throughput on par with major internet platforms, rather than the lesser throughput associated with other blockchain networks.

To put Elrond’s throughput in perspective, the Bitcoin network can confirm 7 transactions per second on average; the Elrond crypto team claims its network can handle 15,000 TPS at a cost of only one cent per transaction.

Elrond accomplishes this extraordinary network efficiency by combining its unique Secure Proof-of-Stake consensus mechanism, advanced sharding technology, validator network structure, and native Elrond coin eGold in a new way (eGLD).

The Elrond Coin: eGold (eGLD)

While the Elrond Network provides a framework for smart contracts, decentralized apps (dApps), and even complete blockchain protocols to be deployed, its native eGLD token serves as the network’s unit of value.

The eGLD coin serves a variety of purposes on Elrond. It’s utilized for staking and validator awards, as well as payment for transactions and smart contracts, as part of the platform governance process.

The eGLD token, which serves as gas for the entire ecosystem, is used by both platform users and developers to pay for use of the Elrond Network.

Elrond’s proprietary Maiar Wallet, which allows users to send near-instant transactions through the Elrond Network, is compatible with the eGLD coin. In addition to staking, Maiar has borrowing, lending, sending, and receiving features.

The circulation quantity of eGLD coins was 17.2 million as of April 2021, with 55.5 percent of that locked up in the network’s staking rewards system. Over the next ten years, new eGLD coins will be produced in a controlled manner according to smart contracts, until the total quantity of 31.4 million eGLD coins is attained.

The comparatively low quantity of the eGLD coin, as well as its lock-up schedule and high percentage of staked coins, are all regarded strong crypto-economic indicators for Elrond’s long-term viability.

>>Bitcoin Wallet Guide, Reviews and Comparison<<

Secure Proof of Stake (SPoS)

The Elrond Network employs a proprietary Proof-of-Stake (PoS) consensus algorithm known as Secure Proof of Stake (SPoS), which incorporates a number of novel features. A node is a server, smartphone, or computer on the Elrond Network that uses the Elrond node client software to relay and process data across the network.

While all PoS networks involve a system of nodes to send and receive data and build consensus, Elrond employs three distinct types of nodes.

Validators are Elrond’s major nodes, and they’re in charge of processing transactions in exchange for eGLD.

Observers are non-active network nodes that read and relay network data. Observers, unlike Validators, are not required to stake eGLD coins in order to participate in the network and are not compensated for their efforts.

Fishermen are specialist nodes that check or challenge data after validator nodes have processed it. A reward is given to a Fishermen node based on how well it is for identifying harmful players.

The Elrond Network’s SPoS mechanism allows for near-instantaneous randomized validator selection, which is made possible via blockchain sharding – the practice of dividing down a large blockchain into smaller, more manageable data sets for maximum performance.

Validator nodes are then assigned to these individual shards at random so that each validator just needs to confirm the data on its own shard, rather than the entire blockchain’s data.

SPoS additionally employs Boneh–Lynn–Shacham (BLS) multi-signature technology to select nodes within each shard at random, completing validator selection in less than 100 milliseconds. This is regarded as extremely fast for a blockchain network, and it is the type of speed for which the term “internet-scale” is reserved.

>>Cryptocurrency Trading Guide for Beginners<<

Elrond’s Blockchain Sharding Innovations

By merging three standardized sharding technologies into one balanced high-performance system, the Elrond Network uses Adaptive State Sharding.

Network sharding makes it easier to allocate and map network nodes to specific shards, whereas transaction sharding determines how transactions are mapped to shards. State sharding, on the other hand, is the most difficult — and it is what distinguishes Elrond from its competitors.

The following is how it works: Because transactions frequently include accounts on separate shards, messages must be sent and network states must be updated across shards.

Nodes within network shards are shuffled to ensure security using randomness to improve resiliency to a variety of threats. When nodes are reassigned and must synchronize to new shards, the combination of these two variables causes latency issues.

State sharding requires nodes to store only a fraction of the total network information, rather than the complete state so that when they are relocated to a new shard, they only have to download a portion of the new state, resulting in a considerably faster synchronization process.

Elrond also makes use of a Metachain to serve as a coordination mechanism for the network’s many shards.

The Metachain runs on its own unique shard, which allows it to communicate directly with all other shards in order to coordinate the cross-shard actions that make the Elrond crypto ecosystem possible.

Smart Contracts on Elrond Crypto Network

The Elrond Virtual Machine, also known as the Arwen WASM Virtual Machine, was created primarily for the purpose of developing smart contracts on the Elrond platform. It works with Rust, C/C++, C#, Go, Typescript, and many more programming languages that can generate Web Assembly smart contracts.

Because of Elrond’s advanced Rust smart contract infrastructure, developers can program in familiar languages, however, Rust is the favored language.

The Elrond Virtual Machine’s (VM) smart contract engine is designed to be fully compliant with the Ethereum Virtual Machine (EVM), allowing Ethereum smart contracts to run smoothly on the Arwen WASM VM.

The Arwen WASM VM is also designed to achieve full interoperability between external blockchain systems, allowing the exchange of value and other data between multiple blockchain protocols.

>>The Best Bitcoin and Cryptocurrency Exchange Guide<<

Elrond Use Cases

Elrond is also designed to enable non-fungible tokens (NFTs) and includes a Smart Accounts NFT architecture. Smart Accounts behave similarly to other blockchain accounts, but with the added benefit of key-value data storage at the account level.

They allow users to save a huge amount of data in their Elrond Network account, including emails, bitcoin addresses, private identification data, health data, proof of citizenship, Know-Your-Customer (KYC) information, and more.

Furthermore, Smart Accounts are designed to directly store and prove ownership of digital assets without the usage of the Elrond VM, resulting in less bloating on the smart contracts that regulate user funds.

Elrond is preparing to launch Elrond Standard Digital Tokens (ESDTs), which are tokenized stablecoins, synthetic assets, and fiat currencies that will run on the Elrond Network but will not require smart contract capability. Elrond’s scalability and speed help prepare the network for the Internet of Things capabilities (IoT).

Today’s businesses looking into blockchain networks as a business solution are looking for platforms that can meet their most pressing requirements.

Elrond Network is a blockchain protocol that is efficient, scalable, developer-friendly, and economical, and it could meet many of the needs of large-scale organizations.

Elrond is delivering solutions for data sharing, IoT, financial services, and many more applications, with a growing list of enterprise blockchain clients throughout the world, demonstrating that user adoption of the Elrond Network is growing stronger.

Website

Dogecoin (DOGE): The Cultural Significance of a Joke Cryptocurrency

EOSIO: Scalability and Enterprise Applications

By Elizabeth Zwicky, anti-spam engineer at Yahoo

At Yahoo, we throw around the term "at internet scale" fairly often. As in, "that design doesn't work at internet scale," or "interesting idea, but not at internet scale." This is what we call “the b problem," which new employees typically respond to with "did you just say 'billion' with a 'b'?" Yes, yes we did. And at internet scale, or in the billions-with-a-b range, life is different. Here are three of the many reasons why dealing with email abuse is unique.

At internet scale, 99% is an F

99% correct should be good, right? In most cases, yes. Except Yahoo Mail has 250 million monthly active users. If we assume more than 100 million people use Yahoo Mail daily, with just a 1% failure rate, you've just gotten it wrong for a million people. That's the entire population of Montana, San Francisco, Calgary, Prague, Glasgow, Milan or Adelaide.

If those 100 million daily users get only 10 mail messages a day, that's a billion incoming mail messages. Now you need 99.9% correct to get it wrong only a million times. Some good news: statistically, you get it wrong with some of the same people twice, lowering the total number of users you’ve annoyed. The bad news: 800,000 angry people is not actually all that much better than a million.

At Yahoo, we measure incoming attempts to send mail to our users in the 10s of billions; that’s 30 to 40 billion emails a day, depending on the day. Getting things right 99% means that’s 300 to 400 million mistakes in a single day. Now, you are no longer creating angry groups of users the size of international cities or American states; you've got major countries you’re aggravating.

When it comes to deciding which accounts to deactivate, we have a 99.999% precision, which means we decide to deactivate a good account for less than one in a million accounts we examine. That one-in-a-million chance? With 100 million daily users, that happens 100 times a day if it's calculated per user.

So 99% won't do at all; 99.9% probably isn't a passing grade; and 99.99% correct is what it takes to be doing OK on most of our systems. We're confident that our accuracy rate of keeping spam out of the inbox is above 99.9%, and we keep adding new ways to make it better.

At internet scale, weird is the new normal

Statistics tell us that to get to 99.99%, we need to be about four standard deviations above (and below) the mean. Just how weird is it to be four standard deviations from the mean? An American man who's four standard deviations above the norm for height is 7 foot 1 inch; four standard deviations below is 4 foot 10 inches. If you want to cover 99.99%, you can't decide that anybody under 5 feet tall is probably a child, and shouldn't be let in, or that a 6 foot 5 inch hallway is good enough.

This makes identifying abuse from perfectly acceptable usage unusually hard. For instance, take an email with the subject line “URGENT!! LIVES AT STAKE!” sent to a large group of people, almost none of whom respond. You would be forgiven for thinking – as one of our algorithms wanted to – that this is from a bad person, trying to swindle strangers out of money. But it's also exactly what you might say if you were trying to find a new home for stray pets. The subject line and behavior is just as characteristic of someone working for an animal rescue shelter as it is of fraud.

We have to care about fraud, but we also have to care about all the corner cases out there; the banks and governments who use our services, the people who create new accounts for online dating. As a result, abnormal doesn’t automatically mean you need to flag it and keep it out.

At internet scale, you're not paranoid; everybody is out to get you.

Everything we do gets abused. It's not just that people abuse our systems for sending mail; they abuse our systems for reporting bad mail. They falsely report other people's accounts as misbehaving. Then, when we get good at noticing that, they turn the spiral another turn and forge false reports. Person A will attempt to make it look like person B is complaining about person C, where A, B, and C don't like each other (they might be anything from kids fighting to competing cybercriminals). Their ideal outcome is that we decide either that C is bad (because B complained) or that B is bad (because the complaint against C is unfounded). If we pick either of those options, A wins. Sometimes the right answer is to punish A; sometimes the right answer is to move on without doing anything to any account; and sometimes the right answer is to punish them all. And no matter what we decide, we lose the time it took to untangle the situation.

At the same time, because *everybody* is targeting us, and weird coincidences are normal, most of what goes wrong is closer to slapstick than to thriller material. Yes, serious and talented criminals attack us. But criminals aren't magicians; they have the same competence ranges as anybody else. Most of them are not very good at what they do. And in a system where improbable things happen often, we spend a lot of time trying to determine whether something is a subtle attack, an accident, or just stupidity. I've spent hours on questions like "is this a clever malicious website or a terrible mistake by a real marketing department?"

The joys of internet scale