Daemon_Y
192 posts
Daemon_Y
@YongDaemon
Work, Crypto, NFT, Crypto, Work
Somnis Rumble Rush is launching on Immutable. A space where dreamlands converge.
Quick, Puffy needs you! Carrots have invaded our Ethereum waters, posing a threat to its decentralization. But no sweat – Puffy’s got this, munching on those carrots like a champ! 🐡🥕 Yes, Puffy's Crunchy Carrot Quest starts... TODAY! Find out more here👇
#KASPA ALERT $KAS by @KaspaCurrency is finally available on #CoinRabbit! Now you can HODL $KAS, while using your funds & optimise your taxes with the fastest crypto lending solution on the market. To use #KAS for loans: tw.coinrabbit.io/cash
Hi $kas, time for a Rusty Kaspa update. The huge success of the first testnet11 run allowed isolating many long-term issues that will make the implementation even better, while not discovering any killer bugs that require immediate attention. This provides the leisure to pursue more long-term goals before initiating the next round of online testing. Among these efforts are: 1. Improving the peer distribution of the peer-to-peer layer. This includes improving the way public node addresses are distributed, as well as making the peer selection out of the set of known peers more robust. @coderofstuff_ -- already familiar for implementing the Ledger integration -- and @averagecatdog (a.k.a jwj in Discord) implemented the approach (already present in the Bitcoin core client) that the class B from which each peer is taken is uniformly random, making it harder to eclipse a node (as this would require the adversary to have nodes in many different address pools). This contribution has already been merged into the code, see the following PRs for details: github.com/kaspanet/rusty… github.com/kaspanet/rusty… 2. Improving the IBD (initial block distribution) process. IBDing with high block rates is subtle, and some less-than-optimal design choices made in the past seems to be less appropriate for 10BPS. This is an ongoing effort by @MichaelSuttonIL . One major correction has already been implemented, you can read the details here: github.com/kaspanet/rusty… 3. Another new contributor, Maxim Biryukov (who has no Twitter account, which is probably the smart choice, but check out his LinkedIn: linkedin.com/in/maxim-biryu…), started working on a profiling infrastructure. Profiling is the practice of scrutinizing a program while it is running to obtain knowledge about its performance (in contrast to debugging, which is the practice of scrutinizing the code). Many of the profiling services generically available in the Go language have no parallels in the Rust language (whose philosophy is that a profiling infrastructure should be tailor-made for the codebase). Maxim is working on utilities that will provide much more detailed data on crucial performance benchmarks, mainly disk I/O pressure and memory and CPU consumption. 4. Integration tests are automated processes that check the correctness of the application as a whole. Unlike unit tests, that verify the correctness of small parts of the code individually, integration tests verify that the behavior of the entire codebase on a large task is as expected. Currently, there aren't enough integration tests implemented on Rusty Kaspa. @OriNewman stepped up to implement integration tests of many functionalities including mining, RPC, P2P and more. 5. Refactoring the mempool: the mempool is the component of the node that stores transactions before they are included in the blockchain. Currently the mempool is "monolothic", which means that once one process accesses the mempool (either to read transaction from it to include them in blocks, or to add new transactions to it) all other processes that need to access it are "locked" until it is finished. When operating in high transaction rates, this creates performance degrading stalls. @Tiram_88 is in the final stages of putting together a first version of an asynchronous refactoring of the mempool. 6. Besides increasing performance, another major goal of the Rust rewrite is to provide a much improved WASM interface for developers over Kaspa. @CryptoAspect is in the final stages of releasing an alpha version of a terminal-driven wallet implemented atop the new interface. Once released, this will allow testing and playing around with the new API and stabilizing it. This is an important step towards the KNG wallet that will succeed KDX.
