第三届中国 Rust 开发者大会 Use Rust to Develop the Decentralized Open Data Application Mike Tang daogangtang@gmail.com @daogangtang 2023-06-08 ➔ 裁员 ➔ 互联网格局定型 ➔ 平台倒闭，数据丢失 这是一个什么时代？ 互联网的终局 创业 -> 种子 com/eightfish-org/eightfish EightFish intends to develop the Data-kind Decentralized Application. You can use the traditional Web development coding style to develop a ODA. How to develop Open Data Application https://github.com/eightfish-org/gutp A Case: GUTP MeBlog is a blog platform for users with the features of open-data and ownership. https://github.com/miketang84/meblog A Case: Meblog Connected to GUTP All

of business is one of the essential functions on the cloud, but the HPA Controller of native K8S use only one goroutine to handle the scaling of all businesses in the cluster, and personalization configuration Pod Product2 Pod Product2 Pod Product1 Pod Product1 Pod Product1 Pod Other products cannot use the resources of product2 How to deal with some products occupying quota but not using it Cluster Quota Online Pod Offline Pod Cannot exceed Offline Quota DynamicQuota System Architechture Diagram QA

implicitly  Since C++20, you need to be explicit, so: [=, this]27 Templated Lambda Expressions  Use familiar template syntax with lambda expressions [] (T x) { /* ... */ } [] use dynamic_cast() and typeid  do dynamic memory allocations, new / delete  contain try/catch blocks  Manually use mutex to protect access to smart pointer  Use global non-member atomic operations, e.g. std::atomic_load(), atomic_store(), …  Error-prone, easy to accidently not use global non-member

size_t  Use case: std::vector data{ 11, 22, 33 }; for (auto i = 0, count = data.size(); i < count; ++i) { /* ... */ }  Doesn’t compile: i is deduced as int, but count is size_t  Solution: use uz literal: switch (number_that_is_only_0_1_2_or_3) { case 0: case 2: handle_0_or_2(); break; case 1: handle_1(); break; case 3: handle_3(); break; default: __builtin_unreachable(); }  C++23: standard way to express assumptions  [[assume(expr)]];  Why use assumptions?  For performance Example: int divide_by_32(int x) { __assume(x >= 0); return

ping); } fn ping(msg: &[u8]) -> wapc::CallResult { …. Ok(msg.to_vec()) } waPC Guest Rust waPC • Use wapc_init to register function • Register_function can only be used inside WAPC_init • Register_function host will still run both "wapc_init" and "_start" during intialization waPC Guest Rust waPC • Use wapc_init to register function • _start also can • Util Lib uses wapc_init, users build on top matching regex extern crate wapc_guest as guest; #[macro_use] extern crate wasm_mock_util; #[macro_use] extern crate serde_json; use guest::prelude::*; use wasm_mock_util::*; fn echo_res_json(msg: &[u8]) ->

the latest announcements Take our survey https://aka.ms/cppcon4 Before we begin... • Please use the Q&A tab for questions during the session (typed) • If you want to raise your hand to ask a question straightforward – what’s the point in creating types for these concepts at all? Couldn’t we just use unsigned int for each? Consider our year_month_day example using operator/: year_month_day just use unsigned int for each? While you could just represent these values with unsigned ints or the underlying data, these new calendrical types also provide valuable abstraction. To use weekdays

make interop suck less by… Explicitly state/document argument content. • Instead of long or int, use int64_t or int32_t. • String encoding is not the same as “width”.What isn’t being declared? struct OBJC_EXPORT id objc_msgSend(id self, SEL op, ...); Indicates variadic arguments… but does it in this case? 2018: objc_msgSend() doc – id objc_msgSend(id self, SEL op, ...); 2019+: objc_msgSend() doc – void make interop suck less by… Explicitly state/document argument content. • Instead of long or int, use int64_t or int32_t. • String encoding is not the same as “width”. Explicitly state/document/reference

and Microsoft STL (MSVC) • Explicit mutex is (thankfully) not used by either implementation • Both use the lowest-order bits on the control_block pointer as a lock26 Daniel Anderson -- danielanderson 16-byte atomics. • Generally ubiquitous and widely available. • std::atomic might not work… • Can use boost::atomic instead!  • Used by Anthony Williams’ implementation • Pack the pointer and guarantee that cb is not deleted until after we are finished reading it Sounds like a textbook use case of deferred reclamation33 Daniel Anderson -- danielanderson.net Deferred reclamation: A quick

add_cpu(int n, float* x, float* y) { for (int i = 0; i < n; i++) y[i] = x[i] + y[i]; } TEST_CASE("cppcon-0", "[CUDA]") { int N = 1 << 20; float* x = new float[N]; float* y = new float[N]; add_cpu(N, x, y); delete[] x; delete[] y; } An Even Easier Introduction to CUDA 4 |TEST_CASE("cppcon-1", "[CUDA]") { int N = 1 << 20; float* x; float* y; cudaMallocManaged(&x add_gpu(int n, float* x, float* y) { for (int i = 0; i < n; i++) y[i] = x[i] + y[i]; } TEST_CASE("cppcon-1", "[CUDA]") { // … } An Even Easier Introduction to CUDA 6 |__global__ void add_gpu(int

执行速度 benchmark • 编译后的体积更小 • borsh 序列化后的二进制更精简 轻量级 序列化结果体积对比 Borsh 基本用法 Case Study NEAR 智能合约 Case Study Solana 智能合约 Case Study • non self-describing • 保证序列化后的二进制唯一性和确定性 • 主要序列化规则 Borsh 规范 •