By Adrian | January 22, 2020
Well what a write off of a year so far. Over the Xmas break, I wasn’t able to get in as many hikes as I would like due to the terrible fires we have had over the last few months. Given the air quality has been rated as hazardous, I’ve erred on the side of caution, not wanting to fill my lungs with it. I was however able to write up the 3 that I did.
I’d also started a number of projects which are partially complete. I had started documenting them, they are at various stages, and some I’m just not happy with how I’ve written about them so will rework them. That being said, this is pretty much what I’ve been doing over the past month.
Looking at DynamoDB as a relational database
But DynamoDB is a no-sql database. While this is true, using the power of Partition, Sort keys and secondary tables, you can structure your data in such a way where it become useable in a relational way. I’ve been attempting to map out my Pokemon pricing data (which is in raw csv files) in this way which ive been collecting for the past 18 months. I am also hoping to be able to do it in such a way where I can set it up as a multi-user system and make it fully online with Lambda and a web front end. Thats still a long way away though, and will involve a fairly steep learning curve for me.
Rick Houlihan does a good job at explaining the theory in the YouTube Video - Advanced Design Patterns for DynamoDB
Security Onion is something I’ve played with in the past, even going down the path of buying a small industrial J1900 4 ethernet port PC from Aliexpress at the time. This was a few years ago now, and when the fan packed it in on that unit, I ended up selling it off as parts. I had attempted to stand up Security Onion again on a virtual machine, hit a roadblock, found another squirrel and moved on. I still want to do a series about this. Security Onion is another great free and open source tool which will give you an Intrusion Detection System that can scale for enterprise.
By far, I’ve spent the majority of my time looking at replacing my aging NAS with something a little nerdier. I have already started the build for this using Single Board Computing devices, specifically the ODroid HC2 which combines an Quad Core ARM processor with 2GB RAM and a single SATA connector capable of taking a 3.5” SATA Drive. Now you might be asking, hows a device that can only take 1 drive going to be a NAS? Well, by using 4 of these devices in a clustered file system (GlusterFS). The 4 ODroid devices will be setup with Docker Swarm to make use of the computing power available, and initially I only plan on having 2 HDD’s setup in replication. This will allow me to add another 2 drives later down the track. The nice part about this setup is that if I need more disk or compute, I can add them in 2 ODroid devices at a time.
So thats pretty much what I’ve been up to in the last month or so, plugging away behind the scenes. I hope to get some more content out soon.