The goal of this type of term paper project is to deeply understand how an exist

The goal of this type of term paper project is to deeply understand how an existing problem is solved ( Fragmentation and allocation techniques in DDBMS) and then slightly improve it. Once you pick a problem (basically, pick anything we mentioned in class, or any component of a system), read the literature to understand the different ways it has been solved. This should require doing a literature search by using Google Scholar, surveys, and following citations. You should expect to read 8-10 or more papers, and to summarize them in your related works section. Since the best way to understand something is to implement it, you will then pick the best-in-class approach from your readings and implement it in a system. For a database algorithm or component, DataBasse may be sufficient, or you may choose some other system. You will then benchmark your implementation and report on the results. After coding, debugging, benchmarking, debugging, laughing, crying, debugging the approach, you should be intimately familiar with the problem. At this point, introduce an improvement. This can be a special case for a particular class of queries or data distribution, or exploiting a hardware optimization, or leveraging some special property of a use case. Once you implement the improvement, update your benchmarks and explain the results. The final report should still be the same paper structure. Naturally, the introduction will focus less on novelty, and more on how the system that you modified did not take advantage of your implementation, and the motivation behind your improvement. In contrast, your related work will be a detailed history and account of how this problem has been solved in the past and their pros and cons.