Jan 2011
26

Meeting #16

-

Concurrency, Scalability & Fault-tolerance 2.0 with Akka Actors and STM


-

LagHunter, a Java-based run-time profiler for user-perceived performance of interactive applications


Venue

Where

University of Lugano, Via G. Buffi 13 - Lugano
Informatics Building, Room SI-008


loading map

When

January 26th, 2011

Agenda

18:30 Concurrency, Scalability & Fault-tolerance 2.0 with Akka Actors and STM
Mario Fusco
19:30 LagHunter, a Java-based run-time profiler for user-perceived performance of interactive applications
Milan Jovic
21:00 A truly effective networking idea: Pizza!

Speeches

Concurrency, Scalability & Fault-tolerance 2.0 with Akka Actors and STM

The biggest challenges in contemporary software development are dealing with overloaded systems, manage concurrency effectively and build applications that scale well. The native Java concurrency model based on threads and their synchronization often doesn't fit these needs or does it at the cost of an extremely high complexity. Akka is a framework, written in Scala, but also working in Java, that try to address those concurrency issues with its actor based model. It also allows to create fault-tolerant systems by implementing a let-it-crash policy through its supervised actor hierarchies and to turns the Java heap into a transactional data set with its clojure-like STM.

Akka of mario - Snoopal

LagHunter, a Java-based run-time profiler for user-perceived performance of interactive applications

A new approach is proposed to deal with performance of highly interactive applications. When developers need to improve the performance of their applications, they usually use one of the many existing profilers. These profilers generally capture a profile that represents the execution time spent in each method. The developer can thus focus her optimization efforts on the methods that consume the most time. It is argued that this type of profiling is insufficient for tuning interactive applications, where the perceived performance relates to how the application responds to user events such as mouse clicks and key presses. This approach allows developers to tackle performance at the highest level of abstraction and at the core of any interactive Java application.

Listener Latency Profiling of NetCoffee - Snoopal

Speakers

Mario Fusco, Ex Machina

Mario joined the Java developers community during his studies at university, and has been a Java enthusiast since. With a long track record of successful projects in several industries ranging from media companies to the financial sector, Mario masters all aspects of software design and development. A clear passion for collaborative software development has led Mario to share his efforts with several communities in mainstream open source software. Current interests include the Scala programming language and functional programming, where Mario has recently released code for two projects, namely LambdaJ and Jamme. Mario is JUG Lugano community leader and a keen speaker at Java conferences and seminars in Europe and the United States. Mario lives in Lugano, Switzerland, where he works as Head of IT development at Ex Machina.


Milan Jovic, Usi

PhD student at USI, department of Informatics under supervision of Matthias Hauswirth. My work is mostly in software performance, with particular focus on performance of interactive applications. Milan holds a Bachelor and Masters from University of Nis (Serbia) at Electrical Engineering Faculty.



Registration Form

Please fill out the form to register to this event



Building Map