Collected for instances outside and inside VPC

I have this collectd cluster configured inside VPC. The metrics are sent to a influxdb instance, then visualized through grafana which is behind a google auth proxy for public access.

| grafana <– influxdb <– node with collectd |

The connection between them are not secured since they are already inside VPC. Now I have these legacy instances which are classic EC2 instances that want to have their metrics on Grafana too. I have 2 choices:
Continue reading


Setup your machine to a server

Recently, I wanted to setup a ssh server on my computer so that I can remote to it in case I need something from my school. Bellow are some of my experience.

First, set up static internal ip for the machine.
Since the computer is connected to internet via a router, we need to set it up to be static to the router. We can do it either from the Network Setup of the machine or in the router using network interface’s MAC address
Continue reading

MapReduce and Hadoop, what are they?

Recently, I have been introduced to MapReduce by my professor. I have heard about MapReduce several times before on the name of seminars and events but I didn’t even check the description because my friends who attended those seminar are geeks in mathematics while I don’t really interested in such subject.

So, what is MapReduce and Hadoop? In sort, MapReduce is a model to process a big amount of data in parallel and Hadoop is the most famous, used and efficient (?) implementation of MapReduce.

Continue reading

Broadcast: best-effort, reliable and uniform reliable

Just like the previous post about fairloss, stubborn and reliable channel (here), I want to note down three types of broadcast that are easily confused: best-effort, reliable and uniform reliable.

Let’s assume that the channel we are using for broadcasting is reliable, which mean that the package that is sent will be delivered and no duplication is made, then:

Continue reading

Fairloss link, stubborn link and reliable link

Recently I got a confusion among those definitions above. After some reading and googling, I found a few things that should be noted down for further reference.

A Fairloss link is a link where the probability the message you send is delivered is non-zero. It means that when you send a message, it is not sure that your message will be delivered but you have hope! If you keep re-transmitting the message infinitely, the message will get to the receiver somehow (infinitely). The point that makes confusion here is that the formal definition keeps talking about “infinitely re-transmission” which easily leads to misunderstanding that the channel automatically re-transmits infinitely. If you send a message once, the channel will send it once and it only says that the message that you sent will arrive to destination with a probability P > 0. Then if you send one message infinitely times, it will arrive in someway but if you send it just once, you have to pray!

Continue reading