incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marc Teufel <teufel.m...@googlemail.com>
Subject Is Cassandra oversized for this kind of use case?
Date Fri, 26 Apr 2013 14:17:52 GMT
I hope the Cassandra Community can help me finding a decision.

The project i am working on actually is located in industrial plant,
machines are connected to a server an every 5 minutes i get data from the
machines about its status. We are talking about a production with 100+
machines, so the data amount is very high:

Per Machine every 5th minute one row,
means 12 rows per hour, means roundabout 120 rows per day = 1200+ rows per
day
multiplied by 20 its 240.000 rows per month and 2.880.000 rows per year. I
have to hold
the last 3 years and i must be able to do analytics on this data. in the
end i deal with roundabout 10 Mio Rows (12 columns holding text and numbers
each row)
Okay, its kind of big data is not really  "big data" isn'it  but for me its
a lot data to handle anyway.
Actually i am holding all these data in a oracle database but doing
analytics on so many rows
 is not the good and modern way i think. as the company is successfull they
will grew, means more machines, again more data to handle...

So i thought maybe Big Data technologies are a possible solution for me to
store my data.

Meanwhile i know Apache Hadoop is not the right tool for this kind of thing
because it scales not down.But maybe Cassandra ? This is my question to
you, do you think cassandra is the right store for this kind of data?

I am thinking about 2 Nodes. Maybe virtual.

Let me know what you think. And if Cassandra is not the right tool please
tell me and if you know any please tell me alternatives. Maybe i am already
doing the right thing with storing that much data in oracle database and
maybe one of you is doing the same - if so please let me also know.

Thank you very much.


Web: http://www.teufel.net

Mime
View raw message