my-hs.txt

Storage Manager Write Up
--------------------------------------------------------------------------------
Design Decisions
1) The first design decision I had to make concerned the representation of heap
    files in the code. I chose to use a struct that just contained the file. I
    also decided to keep pg_cnt in the struct because it helped make the code
    more concise.
2)  The second would be the iterator implementation. I chose to store the current
    pid and the current record index in the iterator so I would know where i amt
    within the heapfile.
3)  Decided to use a hashmap to go from container id's to their respective heap
    files because it allows fast look ups and insertions.

Time Spent, Likes and Dislikes
I spent around 30 hours on this project and I loved the new level of understanding
that I gained from working with the filesystem.

Incomplete Parts
Stress Test and Inserts both fail when the amt of bytes inserted exceed PAGE_SIZE.
My code also fails to meet the final bench. With more time, I would fix both of
these issues by maintining a vector of pages with their sizes for each heap
file. This way I would be able to find the proper page for insertion without
having to pull from disk again each time a page is to small.