Meet the new “High-Performance Blockchain”, Solana

A new standard in Blockchain scalability with up to 710k transactions/second

Abstract

Scalability has recently presented itself as one of the most challenging problems in blockchain technology. By proposing a new blockchain infrastructure based on Proof-of-History (PoH), Solana’s main focus is to set a high-level standard in blockchain scalability with up to 710k transactions/second (tps). It seeks to achieve this by encoding passage of time as data. Solana will be a Proof-of-Stake (PoS) consensus network with smart-contract capability.

What is the problem being solved?

“Our goal is to [build a high performance blockchain that will] make it possible to create a fully decentralized exchange that has the performance of Nasdaq.” — Anatoly Yakovenko, CEO of Solana.

As blockchain technology becomes more popular, the number of transactions on a particular blockchain grows in size. Since this will result in slower transaction times and higher transaction fees, it’s easy to see that a problem occurs. This problem becomes painfully visible when comparing it to other traditional systems. In order to compete with some major traditional systems like Visa, Google and Nasdaq, blockchain technology needs to meet transaction throughputs of 65k, 344k and 500k tps respectively. Bitcoin and Ethereum are currently sitting at throughput rates of 7 and 14 tps respectively. In order to compete against major traditional systems, there’s no other way than to revolutionize the way blockchains are processing transactions. This is where Solana steps in.

Core team

The core team consists of 6 people with strong technical, educational and entrepreneurial backgrounds in businesses like Qualcomm, Dropbox, Google and Intel. We’ll introduce them one by one:

Anatoly Yakovenko — CEO: LinkedIn

Anatoly is no stranger to leadership positions since he was in charge of projects within several businesses like Qualcomm, Mesosphere and Dropbox. He has a strong technical background and holds 2 patents for high-performance Operating Systems protocols.

Greg Fitzgerald — CTO: LinkedIn

With over 10 years experience in Qualcomm, Greg has a very strong technical background in a variety of domains, ranging from JavaScript, to Python to C++ and Haskell.

Raj Gokal — COO: LinkedIn

It seems like Raj is the entrepreneurial mind of the business. Graduated from prestigieus Wharton in 2010, he has been gaining knowledgeable experience in financial operations since. He was a venture investor at General Catalyst, started the consumer medical device company Sano which attracted over $20m in investment, and led product management at Omada Health as it grew tenfold.

Eric Williams — Data Science, Token Economics: LinkedIn

Eric studied particle physics at Berkeley until 2004 and received his PhD from Columbia in 2012 while Higgs Hunting at CERN. He completed a postdoc in Medical Physics at Memorial Sloan Kettering Cancer Center and later became the VP of data science & Analytics at Omada Health.

Alan Yu — Strategy, Business Development, Marketing: LinkedIn

After graduating from Berkeley in 2006, Alan started his career at Google. After spending 10 years at Google in sales and marketing and winning several awards during his tenure, he decided to become a co-founder of Solana.

Stephen Akridge — Engineering: LinkedIn

Stephen has a strong technical background with over 10 years of experience in companies like Intel, Qualcomm and Salesforce.

Solana’s all-rounded technical team will serve them well.

Advisors

Solana is getting technical feedback from people who aren’t interested in being publicly mentioned. They are working on putting together a list of the full team and advisors after a legal structure is in place.

Our 2-cents: In the next section, you are going to read about the pre-established partnership of Solana with companies connected to Greg Kidd (advisor and previous CRO of Ripple) and Alka Gupta (advisor of eBay, PayPal, AT&T and more). It would not surprise us if these top-notch business individuals will eventually turn out to be advisors for Solana. Or perhaps, they will bring in some amazing advisors from their vast array of business connections. Again, this is just purely our speculation and may not happen.

Partnerships

GlobalID — https://www.globalid.net/ — GlobalID is an identification project that lets the user securely, privately, and portably control their identity with groups or other providers and public regulators. CEO and Co-Founder, Greg Kidd, is also an initial investor and advisor for Twitter and Square. President and Co-Founder, Alka Gupta, has a long list of advisory and strategic roles within companies such as eBay, PayPal, AT&T and many more.

Yaka Labs — https://hardyaka.com/— Investment Group is also lead by Greg Kidd. Yaka Labs invests in insights that empower individuals toward a better society. They are early investors of Coinbase, Ripple, Filecoin and many other startups.

impressive!

Technology

Proof of History — a clock for blockchain (vs. Sharding)

Blockchain is all about decentralization, scalability and security. Right now Ethereum is lacking scalability. To solve this, Ethereum will be using Sharding. With sharding, there is no need to keep track of the entire blockchain but little chunks. It verifies the transaction with the hash value with other nodes. However, keeping the entire blockchain on every node is more secure and that’s what Solana is willing to do.

Solana uses a new consensus mechanism called Proof of History which is a sequence of computations that can provide a way to cryptographically verify the passage of time between two events.

Proof of work is not efficient because of the different timestamps on different nodes. Solana has a leader (PoH generator) which keeps track of all the records in sequence. Horizontal scaling is achieved via synchronizing the timings of multiple nodes. Output of the current hash is based on input of the previous hash. With the PoH algorithm, all nodes will know the exact time between each operation that will help to verify the correct hash.

In the example above, photograph in the form of SHA256 is being stored in one node. To verify this hash (336) all other nodes need to know the hash value of the photograph and hash of the previous operation and that can be determined by PoH. Modern GPUs have 4000 cores. All these operations can be done and verified simultaneously.

Time to verify a sequence will be:

Total number of hashes / (Hashes per second per core * Number of cores available to verify)

Zilliqa is also using Sharding and their TPS is around 2500. Solana is going for a much better approach with PoH, which is more secure and reliable than dividing the blockchain into multiple chunks.

Proof of Replication (PoRep) — the Avalanche architecture

A fast streaming Proof of Replication is designed for storing a large distributed ledger. To understand the notion of Replication, you should first understand the concepts “Finality” and “Latency”. You can understand finality in blockchains as a state that cannot be altered anymore. The operation is completed and there is no way the system can ever go back to revert this operation. It’s basically a guarantee past transactions can never be changed. However, it should be stated that finality is in many cases probabilistic and depends on the consensus algorithm of the network. For example, in bitcoin’s Proof of Work consensus a transaction is considered as final when it’s confirmed by 6 blocks. This boils down to 99.99% probability of a transaction not being reversed by an attacker if the attacker controls 8% of the hashing power. However, if the attacker owns 49% of the hashing power, this probability is reduced to 4.06%. There are many consensus models nowadays, which are all dealing with latency and finality in their own ways.

Latency or Finality time, in turn, refers to the time it takes from the creation of a transaction until the initial confirmation of it being accepted by the network. A low latency can be advantageous in multiple situations. For example, while paying with cryptocurrency in a supermarket, you don’t want to wait 30 minutes before your payment is accepted by the network.

Replication involves sharing information so as to ensure consistency between redundant resources to improve reliability in the network. It is a critical piece of every blockchain architecture. If you are able to do it quickly, you are able to advertise little finality times. If you can’t, you’re basically obliged to increase the centralization of the network — Think about EOS with 21 nodes.

With PoH, Solana tries to enable fast and and streaming verifications of Proof of Replication. It wants to be a truly decentralized network, while still allowing for the lightning fast finality times that are required to meet up to 710k tps. This is why they have come up with the Avalanche Architecture. Avalanche has the property that the finality time in the network will only increase with the log of the number of nodes. This means it has miniscule differences in finality times while the growth of nodes goes into 100k, 1 million or even 5 million nodes. Theoretically, this would allow for a truly high-performance, decentralized network. You can think about this logarithmic increase as a society in which 10 residents are connected with each other. If one resident would want to deliver a message, he could just say it out loud and the other 9 residents would instantly know what he said. Let’s increase this society to 100 people. This can be interpreted as level 2. It maybe requires 2 shout-outs to circulate the message to everyone. It will take 5 seconds to get the message across to the first 10 residents (level 1), who will then spend another 5 seconds to deliver it in the same way to the other 90 residents (level 2). So eventually, if you continue this way, it would take approximately the same time to get the message across to 10 million residents instead of 1 million.

Replication is not used as a consensus algorithm, but is a useful tool to account for the cost of storing the blockchain history or state at a high availability.

For more technical details about this architecture, we advise you to consult p21 — p26 in the whitepaper or have a look at this youtube video.

Performant System Architecture

The network is arranged for maximum throughput and high availability. Solana’s smart contracts bytecode is based on Berkeley Packet Filter designed for quick execution.

Consensus independent from transaction flow

Transactions flow through the network independent of consensus and are limited only by hardware capabilities.

MVP

Solana has released its MVP, everyone is invited to take a look at their github activity and run their demo. They claim their testnet is currently performing at 250k tps, peaking at 400k tps. It will be interesting to see this claim confirmed.

Roadmap

It would be nice to see a little bit more detail on the road map even though all essentials items are listed. Now, it’s up to the team to deliver these results in time.

Competition/risk

Solana isn’t the only project that seeks to be the gold standard of high-throughput blockchains. Recently, many projects have popped-up with the same focus and are all impressive numbers to claim the title of World’s fastest blockchain. During early April, Zilliqa claimed to have a throughput of 2k tps during an internal stress test. At the time of writing, Constellation Labs, Credits, Quarkchain and Edenchain are some great examples which had surpassed this number. Edenchain even claims it can theoretically reach unlimited scalability. As previously mentioned, it will be interesting to see which of these companies have the capabilities to put their money where their mouth is when all claims are reviewed.

On top of this, Solana’s whitepaper is mainly focused on technical aspects. We’re curious at how Solana’s impressive technical team will take care of the business side of things: choosing a strategy and building Solana around it, dealing with competition, choosing a target market, teaming up with the right strategic partners, building a strong community, deciding on its token metrics, etc. Knowing their community has quickly grown to over 10k members in less than a month, is a pretty damn good sign.

Conclusion

Solana definitely has the potential to be a big player amongst other high-performers within the blockchain space. 710k tps is an impressive number, especially when the testnet is already running at 250k tps. However, we’d like to see an official confirmation from an independent reviewer before believing such incredible claims. Transaction rates will be published again as the network scales ten-fold from 10 to 100 to 1000 to 10000 nodes. The core team is very impressive. A total of 6 team members remains on the low side to compete for the crown of high-performance blockchains. It will be interesting to see who they will add to such a strong team. Furthermore, the team seems very technical. We’re curious to see how Solana will handle business-related decisions over the next few months.

Thank you for your time.


Please share, clap and follow us for more posts on upcoming Blockchain Innovations.