Interesting article at GigaOm: http://bit.ly/OINpfr I won’t repeat the main
points – but basically it says that since Hadoop is disk/ETL/batch based it
won’t fit for real time processing of frequently changing data. Author
correctly points out that real time processing (i.e. perceptual real time
meaning sub-second to few seconds response time) is becoming a HUGE trend
that’s impossible to ignore. He points to Google that moved away from
Hadoop MapReduce-like approach towards massively distributed in-memory
platform for its various projects like Precolator and Dremel…
So, What’s New?!
The widespread confusion about Hadoop’s role and its applicability is
becoming alarming… Hadoop was never designed to process anything in real
time or process live streaming data or process anything that’s rapidly
changing. Hadoop’s core is HDFS technology – a highly scalable
As increasing numbers of industry leading companies look to new technologies
to speed their real-time services and Big Data applications, GridGain
Systems, leading developer of scalable, distributed in-memory compute and
data grid technologies for real-time data processing, is strongly positioned
to serve the demand.
This week, GridGain announced the release of an entirely new suite of
solution-focused products built around the company’s leading in-memory and
world’s fastest MapReduce technologies, expanded enterprise consulting and
other professional services, advanced DevOps m... (more)
We are happy to announce that GridGain 4.2 is released!
This release includes several new exciting feature as well as the host of
performance optimizations that we’ve included. This release is 100%
backward compatible with 4.x product line and we recommend anyone on 4.x
version to update as soon as possible.
Now – let’s talk about new features…
In GridGain 4.2 we’ve introduced support for delayed preloading. Dmitriy
Setrakyan wrote an excellent blog detailing this new functionality.
Essentially, whenever a new node joins the grid or an existing node leaves
GridGain will be hosting a webinar Thursday, July 26, 2012 at 3pm EST / 12pm
PST during which GridGain’s CTO, Dmitriy Setrakyan, will be live coding
GridGain examples in Java.
Dmitriy will cover the following examples:
How to distribute simple units of work to the grid. Collocation of
computation and data. A full Streaming MapReduce example that performs SQL
queries on streaming in-memory data.
This is a fantastic opportunity to see how easy it is to get started with
GridGain, so register now and join us for “Live Coding In-memory Bid Data
One of the features in GridGain’s In-Memory Data Platform that often goes
unspoken for is ability to launch multiple GridGain nodes in the single JVM.
Now, as trivial as it sounds… can you start multiple JBoss or WebLogic or
Infinisnap or Gigaspaces or Coherence or (gulp) Hadoop 100% independent
runtimes in the single JVM? The answer is no. Even for a simple test run
you’ll have to start multiple instances on your computer (or on multiple
computers), and debug this via remotely connected debugger, different log
windows, different configurations, etc. In one word – awkward…
Not so... (more)