DEVELOPMENT OF LESSLESS AND EFFICIENT DATA STORAGE METHODS FOR IOT
Data management is a significant challenge in handling IoT data. Data must be stored efficiently and queried quickly. While some research has addressed this issue, most of the results are lossy, which can lead to important data being lost if not handled carefully. Therefore, a lossless data storage...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/85478 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Data management is a significant challenge in handling IoT data. Data must be stored efficiently and queried quickly. While some research has addressed this issue, most of the results are lossy, which can lead to important data being lost if not handled carefully. Therefore, a lossless data storage method is needed. Existing lossless methods include horizontal partitioning, data compression, and the Sawalha method. However, none of these methods have successfully addressed all the problems associated with IoT data storage.
To address this issue, a new data storage method has been developed that aims to store data in a smaller size, speed up queries, and ensure losslessness. The proposed method relies on merging several pieces of data into a single record to reduce size and improve data retrieval speed.
To test the proposed method, experiments were conducted on MySQL, PostgreSQL, and MongoDB databases. The results show that the proposed method successfully reduced the data size by up to 95% for integer data. It also improved query speed by up to 99% for most tested queries. However, there is one type of query—aggregation queries without object identity and range—that caused a significant slowdown in query speed with the proposed method. |
|---|