Hazelcast vs Spark: Do they both aim to solve the same problems? If not, how do they differ in the value they deliver?

1,786 views
Skip to first unread message

solo...@gmail.com

unread,
Sep 28, 2013, 7:37:03 PM9/28/13
to haze...@googlegroups.com
I have a project where I need to integrate data in real-time and manage the results in-memory. The aim is to provide a place where fresh, integrated information can be retrieved by and presented to end users quickly.

I looked at a data virtualization approach, but from what I've read, that is more about providing a unified abstraction through which to access a variety of datastores (correct me if I'm wrong).

So I'm looking at either Hazelcast or Spark as a potential place to park post-processed data to serve up, and I'm trying to first understand the particular problems they're trying to solve and which one would be better suited to my project.

Any insight, opinion, or experience is appreciated.

Sol

Christoph Engelbert

unread,
Sep 30, 2013, 1:32:42 PM9/30/13
to haze...@googlegroups.com
Hi Sol,

what Spark are you talking about?

Spark MiniWeb Framework - http://www.sparkjava.com/

My guess would be Apache Spark, am I right?

Chris

solo...@gmail.com

unread,
Sep 30, 2013, 4:50:36 PM9/30/13
to haze...@googlegroups.com
Hey Chris:

Yes, Apache Spark.

Sol

Christoph Engelbert

unread,
Oct 1, 2013, 4:12:49 AM10/1/13
to haze...@googlegroups.com
To be honest I haven't had a llok at it but I will do it until this weekend. Hopefully I'll can give you a good answer :)

alaa...@gmail.com

unread,
Feb 20, 2015, 11:01:33 AM2/20/15
to haze...@googlegroups.com
I have the same question, but I think the difference is that Hazelcast has all data replicated across all servers , in Hadoop/Spark world the data has the divide and conquer  elements in them, so you have a sub-set of data distribute across the servers. 

Let me know what did you find out ...

Ali Gurbuz

unread,
Feb 24, 2015, 5:55:42 AM2/24/15
to haze...@googlegroups.com
Hi,
"Hazelcast has all data replicated across all servers" is wrong, Hazelcast divide the data into partitions and those partitions are distributed to the cluster. Please take a look at 'How the Data is Partitioned in Hazelcast'

--
You received this message because you are subscribed to the Google Groups "Hazelcast" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hazelcast+...@googlegroups.com.
To post to this group, send email to haze...@googlegroups.com.
Visit this group at http://groups.google.com/group/hazelcast.
To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/a329957a-0df6-43dd-9c8e-5cc6a951238c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--

Ali Gurbuz
Core Team Lead

Mahir İz Cad. No:35, Altunizade, İstanbul
a...@hazelcast.com 
+90 507 857 7815
@aligurbuz
Reply all
Reply to author
Forward
0 new messages