Cassandra on a Raspberry Pi. Why ?
Cassandra on a Raspberry Pi.
Why ?
Who Am I ?Andy Cobley
School of Computing
University of Dundee
Twitter: @andycobley
Whats a Raspberry Pi ?
Single chip Linux computer
500 Meg ram
Boots off an SD card
Ethernet port
Pi with pound coin
Here’s oneAnd there is one running in the street area
What’s Cassandra ?
NoSQL database
Important thing is it’s disturbed.
Scales and runs best in a cluster
Data is replicated across nodes
Any node, can answer any query
CassandraCan handle very distributed sites
Data can be in any data center, anywhere.
Typically runs on “consumer class” hardware
The Bad NewsCassandra is designed to be fast, fast at writing, now even faster at reading.
This laptop with one instance of Cassandra will do 12,000 write operations
Raspberry Pi will do 200 !
More bad news !Running a external USB drive is actually worse !
Problem is the USB, the SD card and the Network card all share the same bus
And then there’s Java!
Oracle Java vs OpenJDK
And RaspbianRaspbian is Debian for the PI
Uses the Hard floating point accelerator
Much faster than Debian
Oracle JDK won’t run on it !
The Good News !We’ve forgotten one thing
The Pi cost £25
You can power 4 from USB hub (no need for a power supply on each one
So:
So, have a 64 node computer
for £2000
University of Southhampton
Pi is for teachingSo for £200 we get an 8 node Cassandra cluster
It can be reconfigured, blown away, stress tested and generally abused
We can simulate data racks, data centers and I hope even long network delays.
Our upcoming MSc in Data Science will use these clusters
Pi is discoveryCassandra wouldn’t run on a PI
It does now.
Running it on a Pi shook out some Cassandra bugs
Pi is for fun !Most important, this was pure Geeky Fun
ThanksData Science:
http://www.computing.dundee.ac.uk/study/postgrad/degreedetails.asp?17
Dundee Developer Day and Big Data
17 November in here
http://dun.dddscotland.co.uk/sessions