Date: February 13, 2018
PMP stands for Project Management Professional. It is a project management certification from Project Management Institute (PMI) in the United States. PMP is considered to be one of the prestigious certifications in the project management field.
The basic requirements and process can be found in the PMP handbook on the PMI website.
The PMP certification, along with other project management related certifications offered by PMI, is based on the "A Guide to the Project Management Body of Knowledge (PMBOK Guide)". At the time of writing, PMP is updating from PMBOK Guide 5th edition to 6th edition, effective on March 26, 2018.
I attempted the PMP exam twice and passed on the second time today so it was based on PMBOK Guide 5th edition. However, it should not affect what I am going to say here.

Why did I pursue the PMP?

There are actually many reasons. The main reason is that I really fed up with how my (current time of writing) company does project management. The project management office (PMO) seemed useless in my opinion when it really mattered. No "written" requirements and (functional) managers expect team members (me included) to know what to do. Stakeholders report a defect during user acceptance testing (UAT) when clearly it is a change request because it's not in the design specification.
Actually I can go on forever if I wanted to. Anyway long story short, I hope getting PMP can at least make project management (at my current company) more manageable and controllable, even though I'm not in a Project Manager role.

How did I prepare for the PMP?

I started in March 2017 when I attended a PMP bootcamp. After that, I submitted the application in early April and the one year time limit began. I was lucky that I was not selected for audit.
Before the bootcamp, I knew something about project management like triple constraints (scope, time, cost) but honestly that's about it.
Throughout my preparation, besides the course notes, I used 3 resources mainly:
I used the HFPMP3e as the main text so to speak, since I was already familiar with Head First style from my previous experiences in other technology certifications.
I used the PMBOK5e as a reference text skimming it mostly. Lastly I used RMC8e as a review text ensuring my understanding was aligned with the HFPMP3e text.
I did the readings, made some notes, and did some practice questions. As I progressed, I did like 2 full-length (4-hour) practice exams. All that took like 2 or 3 months, I recalled, since I already scheduled the exam back in April.
D-day came. As exam progressed, I had that feeling in the back of my mind it wasn't good. 4 hours later or when I hit that "End" button, boom, got a "Fail". How bad? Didn't know because there was no score or bar or anything like it on the report, just "Fail".
Okay, back to square one. At that time (Q2 2017), PMI already announced that the PMP will be updating in Q1 2018, but didn't have an exact date. So my goal then was to get it done (pass) by the end of 2017. So I scheduled my second attempt for December 2017.
Since D-day, I actually didn't study or practise for PMP for nearly like 3 to 4 months. Fast forward to like September 2017, with about 2 months to D-day again, I started preparing.
As time passed, I felt I wasn't ready and postponed the exam to February 2018, since PMI had announced the exact date for the new exam (March 26, 2018).
During my second round of preparation, I did things a bit differently:
On the (second) D-day, I felt a bit different from my first attempt. As the exam progressed, I actually was able to eliminate answers down to one in the early questions. The same feeling continued as time ticked away. By the time I was done the first round, I recalled I marked like 15 questions or so that need reviewing. Anyway, I reviewed all questions (clicking "Next" mostly). The third review really was those "marked" questions. There was sufficient time to tackle all 15 or so questions if I wanted to. But after some estimation (on that passing score), I tried tackling a few more. Some of them were calculations and the formulas were unclear so I screwed those. I hit the "End" button and a sign of relief, a "Pass".
Regarding how I estimated whether I passed or not, I used the following. Assume 15 "marked" questions were all wrong:
  1. Double the number of marked questions (15 x 2 = 30)
  2. Less that from 175 (175 - 30 = 145)
  3. Divide that by 175 (145 / 175 = 0.8285)
  4. If the quotient is above say 0.75 (higher is better), you are good
Why divide by 175? Because only 175 questions would be scored, the other 25 are trial questions.
The strategies I used during the exam were:
The notes I made during my preparation can be found here.