InfoQ Homepage Performance & Scalability Content on InfoQ
-
Phusion Passenger/mod_rails makes Rails deployment easy
Phusion Passenger/mod_rails makes deployment of Rails apps simple. The Apache configuration is handled by a script and re-deployment is a single 'touch' away. We talked to the creators of Phusion Passenger who also experiment with a modified Ruby Garbage Collector to share memory across address space borders.
-
Heroku and Morph AppSpaces: two new solutions to Rails hosting
Heroku and Morph Labs are Ruby on Rails hosting providers, offering a complete stack of software and easy to use interfaces to get your applications up and running in a few minutes. We talked to both parties to find out more about their offers.
-
WebSphere Updates: sMash, eXtreme Scale, Virtual Enterprise, Business Events
At IBM IMPACT this week, IBM announced a several new and re-randed upgraded products dealing with virtualization (Virtual Enterprise), clustering & caching (eXtreme Scale), complex event processing (Business Events), and RESTful web apps (sMash). InfoQ spoke to various execs and product managers to find out more.
-
JRuby 1.1 released with major performance improvements
JRuby 1.1 has been released, bringing massive performance increases due to the new JIT, a new Regex engine and other improvements. InfoQ talked to Ola Bini and Charles Nutter about the changes in the new release and the future directions of the project.
-
Are Special Purpose Chips an Answer to the Multicore Crisis?
Adapting to multicore paradigm is one of the emerging challenges in the software industry. What if the solution lays at the hardware level? Bob Warfield suggests that creation of chips optimized for running specific virtual machines could reduce the gap between the potential performance of processors and the actual capacity of software to take advantage from it.
-
Microsoft, Intel to invest $20M in parallel computing
Microsoft and Intel have recently announced a $20 million joint investment into parallel computing over the next 5 years.
-
Using JRuby to generate Code for the JVM
While JRuby's performance keeps increasing, there are still algorithms that are faster if implemented in Java. We look at different approaches to solve this: RubyInline for JRuby, generating bytecode with a JRuby DSL and a new subset of Ruby called Duby.
-
Performance Problems Continue for VS 2008
Like its predecessors, Visual Studio 2008 continues to have performance issues. Unlike VS 2003 and 2005, Microsoft is actively working on performance patches, if only you know where to look.
-
Parallelism with Fork/Join in Java 7
As the number of processor cores available on modern hardware increases, it's becoming ever more important for developers to develop in ways that take advantage of the new hardware. The Fork/Join library in Java 7 helps solve this problem.
-
RadRails goes 1.0 - adds Profiler, CallGraph Analyzer, Rails Shell, etc.
RadRails 1.0, part of the Aptana IDE, has been released. Next to the powerful refactoring capabilities, it adds profiling tools and GUIs for Ruby, fast jruby-debug support for JRuby, and more. We talked to Christopher Williams of RDT and Aptana about RadRails 1.0.
-
XSLT Profiling in VS 2008
Microsoft's XML Team has released a preview of their XSLT profiler for VS 2008. Unfortunately, only Team System users get to try it out.
-
POJO Messaging Architecture with Terracotta
Mark Turansky detailed his implementation of a message bus architecture using Terracotta and Java 5. Instead of using an MQ or JMS based deployment, Mark took advantage of the Terracotta architecture to create his POJO message bus. This allowed for a clean, simple, and inexpensive infrastructure solution to his message needs.
-
MacRuby - Ruby 1.9 ported to Objective-C
A new project called MacRuby aims to improve Ruby on MacOS X by using the Objective-C runtime and Garbage Collector to improve Cocoa support and speed. To get an idea of how MacRuby works, we talked to Laurent Sansonetti of the MacRuby team.
-
Generic versus User Specific Data Streams for Scalable Web Sites
Describes an approach to scaling web applications by partitioning data according to what is generic and what is user specific. The generic data streams can then take advantage of horizontal scaling and the power of caching.
-
Trading Consistency for Scalability in Distributed Architectures
eBay's Dan Pritchard and Amazon's Werner Vogels talk about the necessary trade-offs to achieve appropriate network partitioning tolerance for large distributed systems.