## +23 ## Distributed Ranges: A Model for Building Distributed Data Structures, Algorithms, and Views ## BENJAMIN BROCK ## Notices and Disclaimers For notices, disclaimers, and details about performance 33cc9ed6c477/p6_1.jpg) ## Project Goals - Offer high-level, standard C++ distributed data structures - Support distributed algorithms Achieve high performance for both multi-GPU, NUMA, and multi-node 477/p7_1.jpg) ## Outline - Background (Ranges, Parallelism, Distributed Data Structures) - Distributed Ranges (Concepts) - Implementation (Algorithms and views) - Complex Data Structures

## A Multithreaded, Transaction-Based Locking Strategy for Containers Bob Steagall CppCon 2020 KEWB COMPUTING ## Overview • Sharing a container among multiple threads • A motivating problem • Some modification of an update group is treated as a single transaction, which either succeeds completely, or fails completely - Consistent: Each transaction can only bring the update group (and the enclosing enclosing container) from one valid state to another, maintaining all invariants - Isolated: Each transaction must ensure that concurrent execution of other transactions leaves its update group (and the container)

## Using MySQL for Distributed Database Architectures Peter Zaitsev CEO, Percona PingCAP Infra Meetup, Shanghai, China, May 26, 2018 ## About Percona Solutions for your success with MySQL, MariaDB, Internet companies and enterprises ## Presentation ## Cover Basics ## Why Going Distributed ## How to do it ## Distributed ? ## MySQL Deployment on More than one System ## Modern MySQL Scalability ![ 28.000.000 Daily Active Users Possible 15M of Daily Active Users counting time of day skew ## Distributed Systems Tend To be More Complicated to Develop Against More Complicated to Operate ## Have Additional

## 67 -328 Building Distributed Applications WebSockets ### HTTP / AJAX / WebSockets / Socket.io • Recall that HTTP is request/response protocol • AJAX allows you to do HTTP request/response within

## +21 ## PGAS in C++: A Portable Abstraction for Distributed Data Structures ## BENJAMIN BROCK 20 21 October 24-29 ## About Me - PhD candidate at Berkeley - Advised by Kathy Yelick and Aydın Buluç program for a supercomputer? Introduce PGAS Model, RDMA Building Remote Pointer Types Building Distributed Data Structures Extending to GPUs This Talk Background: how do we write a program for a supercomputer supercomputer? Introduce PGAS Model, RDMA Building Remote Pointer Types Building Distributed Data Structures Extending to GPUs This Talk Background: how do we write a program for a supercomputer

ShardingSphere-Sidecar(TODO) 1.1.4 Hybrid Architecture 1.2 Features 1.2.1 Data Sharding 1.2.2 Distributed Transaction 1.2.3 Database Governance 2 Quick Start 2.1 ShardingSphere-JDBC 2.1.1 Import Maven Core Concept 11 Overview 11 SQL 11 Sharding 13 Configuration 14 Inline Expression 16 Distributed Primary Key 19 Hint Sharding Route 21 3.1.5 Guide to Kernel 22 SQL Parsing 23 Query Optimization Background 52 SQL 52 Pagination 57 2 Distributed Transaction 59 3.2.1 Background 59 Local Transaction 59 2PC Transaction 60 BASE Transaction 60 3.2.2 Challenge 61 3.2.3 Goal 61 3.2

DataSourceTransactionManager closes JDBC connection on doBegin failure but leaves it attached to transaction object (SPR-12280) fixed webAsync responses force concurrentResult.toString() call (SPR-12253) fixed MBeanExportConfiguration.SpecificPlatform issue (SPR-11877) allow manually registered transaction manager to be recognized by qualifiedBeanOfType (SPR-11915) fixed misleading exception message attributes (SPR-9553) * JmsTemplate uses configured receiveTimeout if shorter than remaining transaction timeout (SPR-10109) * added MappingJackson2MessageConverter for JMS (SPR-10099) * JDBC parameter

well * fixed and updated reference documentation on JMS listener container configuration and transaction participation Package org.springframework.aop * Cglib2AopProxy uses weak references for advised the exception message * fixed Hibernate/JpaTransactionManager to clean up properly in case of transaction begin failure on pre-bound resource * revised JpaTransactionManager to convert JPA RollbackExceptions Creator's "joinTransaction()" checks whether there is an externally managed local transaction * ExtendedEntityManagerCreator's EntityManager proxies are serializable now $ ^{*} $ (

setExecutor method to correctly wrap in a TaskExecutorAdapter * @Transactional qualifiers match against transaction manager definitions in parent contexts as well * TransactionSynchronizationManager eagerly cleans r type * JmsResourceHolder does not ignore IllegalStateException from locally synchronized transaction * fixed JMS CachingConnectionFactory to correctly cache a producer without fixed destination as DefaultJdoDialect supports JDO 3.0 query timeout facility (as supported by DataNucleus 2.1) * transaction names based on method id from most specific method (target class instead of interface) * added

JobDetails in combination with autodetected JobDetailAwareTriggers Package org.springframework.transaction * fixed AbstractPlatformTransactionManager to always execute all synchronizations, even on unexpected JtaTransactionManager calls "afterCompletion" of Spring synchronizations at completion of outermost JTA transaction * JtaTransactionManager throws UnexpectedRollbackException in case of global rollback-only if default data access strategy to JDBC, to highlight the JMX export Package org.springframework.transaction * exposed "hasSavepoint" and "isCompleted" methods on TransactionStatus interface