As a part of the food insecurity early warning system based on local participation, a robust and scalable database service is required. This necessity caused by the large area of services which include 34 provinces, 416 districts, 7,215 sub-districts and 80,534 villages in Indonesia. The abundant number of the expected daily transaction might not be handled properly using the traditional model. In this research, we design, implement, and optimize the NoSQL database to create scalable, dynamic, and flexible database service for the early warning system. The cohesion of the model is then measured, resulting in 5 entities with high cohesion, 16 with moderate cohesion, and 3 with low cohesion. After refactoring, we reduced the number of the low-cohesion entity into one and increased the average cohesion from 0.62 to 0.67. An empirical experiment was conducted to compare the response time before and after the refactoring. As the results, the average response time is decreased from 11.0 ms to 7.99 ms or equal to 1.38 in speedup. The experiment results suggest there is an impact of the logical data model improvement, by increasing their cohesion, to the performance of the NoSQL database.

In this paper, we propose a new style of tower defense game by using the circular placement of the tower. In this game, dubbed Orbital Defense X, the player must protect the earth with a circle of defensive tower circled around it. The game is developed using game development lifecycle from initiation phase through alpha version. To test the quality of the developed game, we measure the game experience using game experience questionnaire. The questionnaire is divided into three: core, social presence, and post-game questionnaire. The results show that male participant gains a significantly higher score than female participants. Overall, the game experience is received better by the male participant in the core and social presence components, with an average score of 2,27 vs 1,92 and 2,15 vs 1,86. However, in post-game experience, the female participant scores higher with 2,15 vs 1,95. The participant that previously played another tower defense game gave score more in the core, social presence, and post-game experience than the participant that haven't played it before. Their score is 1,93 vs 2,37 in core, 1,67 vs 2,39 in social presence, and 2,07 vs 1,91 in post-game experience. The result also suggests for more attention in the challenge regarding the enemies (variety and difficulty) rather than more complex tower management system. The social ranking also suggested by the participant to improve the overall experience of this game.

As a part of the food insecurity early warning system based on local participation, a robust and scalable database service is required. This necessity caused by the large area of services which include 34 provinces, 416 districts, 7,215 sub-districts and 80,534 villages in Indonesia. The abundant number of the expected daily transaction might not be handled properly using the traditional model. In this research, we design, implement, and optimize the NoSQL database to create scalable, dynamic, and flexible database service for the early warning system. The cohesion of the model is then measured, resulting in 5 entities with high cohesion, 16 with moderate cohesion, and 3 with low cohesion. After refactoring, we reduced the number of the low-cohesion entity into one and increased the average cohesion from 0.62 to 0.67. An empirical experiment was conducted to compare the response time before and after the refactoring. As the results, the average response time is decreased from 11.0 ms to 7.99 ms or equal to 1.38 in speedup.The resulting database is then used as a part of database services in our early warning system.