Your client asks you to design a Data Storage scheme for a library data system u

Question

Your client asks you to design a Data Storage scheme for a library data system under the following assumptions:
-Data system contains hundreds of thousands of large data records (consider the range of 600,000 as an example)
-The initial file is available sorted.
-In terms of frequency, you expect many insertions/deletions and queries involving multiple attributes performed each second.
-Queries are sent frequently and in real-time using these field attributes of the record: ISBN number, Title, and Author fields.
-Queries are sent occasionally using these field attributes of the record: Publisher, and Subject.
-The entire system is archived monthly.
-There is no requirement for printing the file in order.
-You may make reasonable assumptions about hardware availability.

Sort types:
+Exchange
+Selection
+Insertion
+Merge
+Bubble
+QuickSort
+Heap
+Binary Tree
+Shell
+Address Calculation
+Straight 2-way Merge
+Natural Merge
+Radix

Things to consider these points in choosing Sort types:
-size of data
-order of data
-distribution of data
-re-use of existing code
-programmer time
-frequency of searching
-how is the data used?
-efficiency of sorts available
-space requirements

Search strategies using Sorted data:
+Sequential (native or brute-force)
+Binary
+Interpolation
+Indexed-Sequential
+Search Trees
+must maintain sorted order when performing Insertions/Deletions.

Search strategies using Non-sorted data:
+Sequential (brute-force)
+Transposition
+Move-to-the-Front (cousin to Transposition)
+Hashing

Things to consider these points in choosing Search strategies:
-size of data
-distribution of data
-re-use of existing code
-programmer time
-frequency of searching
-number of search types
-search strategies exploit the file organization to efficiently find items
-it is common to search for items NOT in file, in order to prevent duplicates.
-Nonsorted items does not necessarily mean unorganized.
?
QUESTION: What do you propose as the data storage scheme, and why? Consider which Sort types and Search strategy algorithms you would use considering the given factors, and why?

?

Explanation / Answer

To storage large amount data the relational database is one of the best way, where we can easily store and retrieve the data.We can also use NoSQL storage technologies prove a key enabler to efficiently analyse large amounts of data and create additional business value.

In terms of sorting large amount of data ,according to me merge sort is the best way to sort the elements because it time complexity even for worst case is o(nlog n) which is lesser than other sorting algorithm.

In searching algorithm Hashing technique is one of the best way beacuse it is based on key,with the help of key can retrieve,insert and update the element in less time.

We can easily prevent the duplicacy in dataset with the help of hashing technique.

Related Questions

Navigate

• Browse (All)
• Browse Y
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.