Skip to main content

CST363 - Learning Journal Week 5

 This learning journal entry is in response to the following article:

https://use-the-index-luke.com/sql/anatomy/slow-indexes

Q: If indexes are supposed to speed up performance of query, what does the author mean by a slow index?

In the article the author refers to a slow index as an index where the database traversing not only the index tree structure but also multiple leaf nodes to find all matches. What this means is that although indexes help the database close in on the requested data without having to look at every node, it can still be slow because it will need to follow multiple branches to ensure it finds all of the matching keys. With leaf nodes possibly having many matches the requested data can span across multiple blocks. 

Slow indexes are a myth due to the mistaken belief that the index lookup only involves tree traversal when in reality is also involves following leaf node chains and fetching the table data where there is no limit to the amount of block access requests that it may need to perform.


Comments

Post a Comment

Popular posts from this blog

CST311 Week 6 Journal

 This week, we developed a deeper understanding of the network control plane and explored routing algorithms such as Dijkstra and Bellman-Ford, focusing on how they determine the most efficient paths for data transfer. We also learned about the SDN control plane, which allows for dynamic network management and routing. In addition to the theory learned this week, we conducted a network lab exploring the use of a basic SDN controller using Mininet and Wireshark. We looked at how Internet Control Message Protocol (ICMP), can be used for network diagnostics, particularly for network troubleshooting with tools like ping and traceroute.
Post a Comment
Read more

CST335 Week 7 Journal Entry

 This week I learned about input/output (I/O) devices and their interactions with the operating system (OS). Two types of devices I learned about were block and character devices. Block devices, like hard drives and SSDs, store data in fixed-size blocks and allow random access, making them ideal for large data transfers. Character devices, such as keyboards and mice, handle data streams one character at a time, supporting sequential access. The OS play a crucial role in managing these devices through device drivers that provide a standard interface to interact with the hardware. Performance characteristics of hard drives, such as seek time, rotational latency, and transfer rates, are pivotal in determining overall system efficiency. By calculating transfer rates for different workloads, I learned how disk access patterns significantly impact performance. Additionally, understanding RAID levels provided insights into how redundancy and striping enhance data reliability and performan...
Post a Comment
Read more

Leetcode Practice Reflection 1.Two Sum

LeetCode Practice Reflection - 1. Two Sum Date:   19MAY25 What I Worked on Today: Practiced doing LeetCode Problem 1.Two Sum. What I Learned: - How to use the enumerate() function - How to optimize possible solutions after considering brute force methods - How to declare a new dictionary (key value pair) Code Snippet I Wrote or Studied: class Solution(object): def twoSum(self, nums, target): seen = {} for i, num in enumerate(nums): compliement = target - num if compliment in seen: return [seen[compliment], i] seen[num] = i In My Own Words: This function takes in a list of numbers and will return the index of the two numbers that add up to the provided target value. The way this program works is that it creates a dictionary called "seen". It will then iterate through the list of numbers and use the enumerate function to keep track of the index and the value of the specific item in the provided list. The for loop will calculate the compliment numbe...
Post a Comment
Read more