SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : 2026 TeoTwawKi ... 2032 Darkest Interregnum
GLD 368.29+0.6%Nov 7 4:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: maceng2 who wrote (178323)9/15/2021 5:14:17 AM
From: TobagoJack   of 217591
 
<<Cryptos>> link to hardware discussed by one of the band. Am learning, or trying to keep up. Used to study computer structures. Never worked with whatever knowledge gained in the field of IT, but am familiar-enough for casual curiosity purpose. Enough vocabulary here not familiar that I needed to google along.

reddit.com

What happened to SOL summer??


I need to re-share this. On one of my previous posts where i compared CSPR to AVAX, someone commented asking why I was focusing on the hardware requirements for validators in the comparison:

reddit.com

We can see now with the disruption to SOL, why you cannot start off too heavy weight in terms of specs required for hardware. Blockchains only grow and so getting the requirements on day 1 just right is more important than most people understand. Have a fancy 2 protocol sharded mechanism to increase throughput is irrelevant when it brings down the entire networks. So to recap...here AGAIN, is our comparison of CASPER VS SOLANA:

ghoststaking.com

Summary:

Network nodes:

The SOL network of 200 physically distinct nodes are all run with GPUs. Where each node requires 256 GB of RAM.

CSPR at present has 100 validators which are able to run on standard servers as the node only required 32 GB of RAM

Commentators have repeatedly highlighted the fact that the cost to run a Solana node is much higher than in other networks.

Decentralization:

SOL’s degree of decentralization is still very much up for debate. To become a validator on Solana, an individual would need to shell out thousands of dollars in hardware. This is in contrast with other blockchains including CSPR where anyone can become a validator for much less.

With regard to the tokens:

60% of SOL are controlled by the project’s founders and the Solana Foundation.

This compares with CSPR which totals 24% (where the team hold 8%, advisors hold 6% and CasperLabs Holdings AG holds 10%).

Throughput:

This we feel is the key to the comparison. SOL is able to achieve 50k tps (transactions per second) utilizing an expensive network of very high powered GPUs across 200 nodes. This compares with CSPR which is currently achieving 2.5k transactions per block (100 WASM deploys per block) by utilizing only 100 low power validator nodes (with a possible increase of x5 throughput being tested at present). The way that SOL achieves the high throughput is by the implementation of its parallel PoH protocol. As this is not part of its security consensus protocol, but a mechanism to simply increase throughput, it can be considered in the same way has many sharding methodologies being researched currently. As we are aware (from the final section of our sharing article here), CSPR is currently researching a proposal on how to multi thread (parallel processing) transactions. To summarize the proposal: CSPR Sharing – Transactions will list all spaces in memory (shards) which they require and each block is given a limited amount of sequential computation time. When a block includes a bunch of transactions instead of specifying a transaction order, it specifies a computation schedule. This schedule uses times from 0 to t where t is our bound on sequential compute time. Each transaction is assigned a sub-interval of this window with the length of the sub-interval given by the computation bound specified by the transaction. Furthermore, the intervals for any two transactions that share a shard are not allowed to overlap. Distinctions could also be made between reads to a shard and writes to a shard and allow several transactions to simultaneously read from the same shard so long as none write to it.

You can see from the above that it is inappropriate to compare the throughput of SOL with that of CSPR as without sharding implemented – the comparison is not like for like. A more direction comparison on throughput should be made post the implementation of sharding on the CSPR network. It should be noted however, that once sharding is implemented, high throughput will be achieved on CSPR without the need to expensive GPU infrastructure.

Security:

Security is one of the most fundamental aspects of blockchain technology, as without sufficient security, no utility can be generated. On this point, I would like to refer the reader to our extensive commentary of the CSPR CBC protocol and Highway protocol. These two protocols in conjunction provide by far the most superior security consensus of any PoS blockchain. No statistical methods are used and security is not affected via conflicting messages. .

Solana’s PoS system relies on a Byzantine Fault Tolerance (BFT) mechanism called Tower Consensus. Tower Consensus leverages PoH as a global source of time before consensus is achieved in order to reduce latency. Any validator node is eligible to be chosen as the PoH leader. If there is any failure detected with the PoH generator, then the validator node with the next highest voting power will be chosen to replace the original leader.

Slashing:

Slashing is implemented within SOL. A malicious vote will remove a validator’s bonded tokens and add them to the mining pool. Slashing also occurs if a vote is cast for an invalid hash generated by the PoH generator.

Slashing will not be implemented in CSPR. Instead malicious validator node behaviour will be punished with severe jail time

Enterprise adoption:

At present there are is no focus on Enterprise adoption with SOL. This is in contrast with CSPR where CasperLabs have aggressively targeted and pivoted towards the running of both a public and private blockchain. Read more here

Future Proof:

CSPR has been designed and engineered such that it is a fully future proof blockchain. It has achieved this by ensuring all components (Execution Engine, Network Engine, Consensus Layer) are all pluggable. This allowed the blockchain to be upgraded with relative ease. In comparison, SOL has taken the stance to create a protocol which is theoretically designed to scale with Moore’s Law, doubling in capacity every two years with improvements in hardware and bandwidth.

Looking to stake CSPR? support our research by staking with us at GHOST. ghoststaking.com We run a:

(1) Low fee powerful node

(2) guarantee of no fee increases ever

(3) live node monitor on our website

follow us on twitter for more publications: twitter.com

join our CSPR price chat tg group: t.me
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext