Richard Kreuter and Kyle Banker on how to avoid classical RDBMS transactional systems by using compensation mechanisms, transactional messaging or transactional procedures.
Cyprien Noel discusses distributed transactional memories along with ObjectFabric, a Java server based on eXtensible Software Transactional Memory, an OS library for concurrent and distributed apps.
Jags Ramnaraya presents SQLFire and how SQL can be used for modern data stores backing online highly scalable applications by using a different consistency model and sharing nothing persistence.
Martin Thompson and Michael Barker talk about building a HPC financial system handling over 100K tps at less than 1ms latency by having a new approach to infrastructure and software.
Bill Burke shows how to use REST to create interfaces to middleware services – messaging, transactions, workflow, security – in order to have RESTful enterprise SOA implementations.
Justin Sheehy explains the principles behind concurrent distributed systems: no global state, no ACID but rather BASE, no RPC but protocols over APIs, prepare for failure, degradation, measurement.
Gregor Hohpe of Google discusses software as connecting services and components, describes the constraints of connected systems design, and presents common design patterns to solve those constraints.
Our application runs over 10,000 sustained transactions per second with a rich model. The key? Modeling state transitions explicitly.
This presentation covers Betfair's efforts, e.g. Flywheel, that enables serving 50,000 low cost transactions per second. This technology has become the basis for the Tradefair financial exchange.
David Chase discusses Fortress, a Fortran-based HPC programming language. Topics include Fortress origins, running on the JVM, work stealing, transactions, continuations and the type system
William Soo and Meeraj Kunnumpurath discuss the Voca transaction processing system, architectural challenges and requirements, Voca's Spring/J2EE architecture, and the future SEPA architecture.
Juergen Hoeller explains the "dark art" of transaction management within high-volume mission-critical Java EE applications backed by Spring 2.