Advent of Code F# – Day 22

You can become a serverless blackbelt. Enrol to my 4-week online workshop Production-Ready Serverless and gain hands-on experience building something from scratch using serverless technologies. At the end of the workshop, you should have a broader view of the challenges you will face as your serverless architecture matures and expands. You should also have a firm grasp on when serverless is a good fit for your system as well as common pitfalls you need to avoid. Sign up now and get 15% discount with the code yanprs15!

ps. look out for all my other solutions for Advent of Code challenges here.


Day 22

See details of the challenge here.

Today’s input looks like this:

root@ebhq-gridcenter# df -h
Filesystem                           Size    Used     Avail    Use%
/dev/grid/node-x0-y0     87T     71T     16T       81%
/dev/grid/node-x0-y1     93T     72T      21T       77%
/dev/grid/node-x0-y2     87T     67T      20T       77%
/dev/grid/node-x0-y3     89T     65T      24T       73%
/dev/grid/node-x0-y4     93T     67T      26T       72%

Of the information presented to us we only need the x & y position of the node, its size and amount of space used. Everything else can either be derived from these or are not needed for this challenge.

For part 1 of the challenge, if we sort the array of nodes by the amount of available space in descending order, then for each of the elements we only need to iterate through the sorted array for as long as the available space is >= the used space of the current node. Hence reducing the average time complexity from O(n2) to O(n * log n).


Part 2

Now that you have a better understanding of the grid, it’s time to get to work.

Your goal is to gain access to the data which begins in the node with y=0 and the highest x (that is, the node in the top-right corner).

For example, suppose you have the following grid:

Filesystem            Size  Used  Avail  Use%
/dev/grid/node-x0-y0   10T    8T     2T   80%
/dev/grid/node-x0-y1   11T    6T     5T   54%
/dev/grid/node-x0-y2   32T   28T     4T   87%
/dev/grid/node-x1-y0    9T    7T     2T   77%
/dev/grid/node-x1-y1    8T    0T     8T    0%
/dev/grid/node-x1-y2   11T    7T     4T   63%
/dev/grid/node-x2-y0   10T    6T     4T   60%
/dev/grid/node-x2-y1    9T    8T     1T   88%
/dev/grid/node-x2-y2    9T    6T     3T   66%

In this example, you have a storage grid 3 nodes wide and 3 nodes tall. The node you can access directly, node-x0-y0, is almost full. The node containing the data you want to access, node-x2-y0 (because it has y=0 and the highest x value), contains 6 terabytes of data – enough to fit on your node, if only you could make enough space to move it there.

Fortunately, node-x1-y1 looks like it has enough free space to enable you to move some of this data around. In fact, it seems like all of the nodes have enough space to hold any node’s data (except node-x0-y2, which is much larger, very full, and not moving any time soon). So, initially, the grid’s capacities and connections look like this:

( 8T/10T) --  7T/ 9T -- [ 6T/10T]
    |           |           |
  6T/11T  --  0T/ 8T --   8T/ 9T
    |           |           |
 28T/32T  --  7T/11T --   6T/ 9T

The node you can access directly is in parentheses; the data you want starts in the node marked by square brackets.

In this example, most of the nodes are interchangable: they’re full enough that no other node’s data would fit, but small enough that their data could be moved around. Let’s draw these nodes as .. The exceptions are the empty node, which we’ll draw as _, and the very large, very full node, which we’ll draw as #. Let’s also draw the goal data as G. Then, it looks like this:

(.) .  G
 .  _  .
 #  .  .

The goal is to move the data in the top right, G, to the node in parentheses. To do this, we can issue some commands to the grid and rearrange the data:

  • Move data from node-y0-x1 to node-y1-x1, leaving node node-y0-x1empty:
    (.) _  G
     .  .  .
     #  .  .
  • Move the goal data from node-y0-x2 to node-y0-x1:
    (.) G  _
     .  .  .
     #  .  .
  • At this point, we’re quite close. However, we have no deletion command, so we have to move some more data around. So, next, we move the data from node-y1-x2 to node-y0-x2:
    (.) G  .
     .  .  _
     #  .  .
  • Move the data from node-y1-x1 to node-y1-x2:
    (.) G  .
     .  _  .
     #  .  .
  • Move the data from node-y1-x0 to node-y1-x1:
    (.) G  .
     _  .  .
     #  .  .
  • Next, we can free up space on our node by moving the data from node-y0-x0 to node-y1-x0:
    (_) G  .
     .  .  .
     #  .  .
  • Finally, we can access the goal data by moving the it from node-y0-x1to node-y0-x0:
    (G) _  .
     .  .  .
     #  .  .

So, after 7 steps, we’ve accessed the data we want. Unfortunately, each of these moves takes time, and we need to be efficient:

What is the fewest number of steps required to move your goal data to node-x0-y0?

If you visualize the grid then the solution becomes pretty clear and one that we can actually work out by hand.

Here, the target node is marked as red, the empty node green (which is initially at position (7, 17)), and the very large, very full nodes are marked as black. We want to move the empty node all the way to (35, 0) so we can start moving the target node at (36, 0) horizontally towards (0, 0).

We can again use the Level Order Tree Traversal approach we have seen so many times this AOC for the first part. First, let’s transform the Node[] into a 2D array first to make our task easier.

The trick for the traversal part of the problem is to avoid passing through the same coordinate more than once, a problem that can be easily rectified with a cache of positions that we have already passed through.

Once the empty node is at (35, 0) we can commence the second part of the problem – to move the data in the target node to (0, 0). Every time we move the data to the empty node on the left, we’ll have to move the empty node in front of it again, which means going down, left, left and then up, so that means a total of 5 moves for every step towards (0, 0). Except, when the data is at (1, 0), which requires only one move.

Which means, there’s a total of 5 * 35 + 1 move left ONCE we have move the empty node to (35, 0).

In case you’re wondering why I left out the “+ 1” on line 33, it’s because the path that’s returned by findPaths includes the starting position (7, 17), so really line 33 should have read let part2 = path.Length – 1 + 35 * 5 + 1



Liked this article? Support me on Patreon and get direct help from me via a private Slack channel or 1-2-1 mentoring.
Subscribe to my newsletter

Hi, I’m Yan. I’m an AWS Serverless Hero and I help companies go faster for less by adopting serverless technologies successfully.

Are you struggling with serverless or need guidance on best practices? Do you want someone to review your architecture and help you avoid costly mistakes down the line? Whatever the case, I’m here to help.

Hire me.

Skill up your serverless game with this hands-on workshop.

My 4-week Production-Ready Serverless online workshop is back!

This course takes you through building a production-ready serverless web application from testing, deployment, security, all the way through to observability. The motivation for this course is to give you hands-on experience building something with serverless technologies while giving you a broader view of the challenges you will face as the architecture matures and expands.

We will start at the basics and give you a firm introduction to Lambda and all the relevant concepts and service features (including the latest announcements in 2020). And then gradually ramping up and cover a wide array of topics such as API security, testing strategies, CI/CD, secret management, and operational best practices for monitoring and troubleshooting.

If you enrol now you can also get 15% OFF with the promo code “yanprs15”.

Enrol now and SAVE 15%.

Check out my new podcast Real-World Serverless where I talk with engineers who are building amazing things with serverless technologies and discuss the real-world use cases and challenges they face. If you’re interested in what people are actually doing with serverless and what it’s really like to be working with serverless day-to-day, then this is the podcast for you.

Check out my new course, Learn you some Lambda best practice for great good! In this course, you will learn best practices for working with AWS Lambda in terms of performance, cost, security, scalability, resilience and observability. We will also cover latest features from re:Invent 2019 such as Provisioned Concurrency and Lambda Destinations. Enrol now and start learning!

Check out my video course, Complete Guide to AWS Step Functions. In this course, we’ll cover everything you need to know to use AWS Step Functions service effectively. There is something for everyone from beginners to more advanced users looking for design patterns and best practices. Enrol now and start learning!

2 thoughts on “Advent of Code F# – Day 22”

  1. Pingback: Advent of Code F# – Day 23 |

  2. Pingback: F# Weekly #52, 2016 – Sergey Tihon's Blog

Comments are closed.