Differences

This shows you the differences between two versions of the page.

--- report [2024/06/17 02:03] – [7.6 Prototype] team1
+++ report [2024/06/19 21:40] (current) – [3.1 Scope] team1
@@ Line 414: / Line 414: @@
 ==== - Scope ====
-In order to be more precise about the scope of the project than in the introduction, from a project management point of view, the following objectives are to be achieved, also illustrated in Figure {{ref>scope}}: \\
+In order to be more precised about the scope of the project than in the introduction, from a project management point of view, the following objectives are to be achieved, also illustrated in Figure {{ref>scope}}: \\
  - Initial research: First of all, initial research has to be done to find out the state of the art. At this point, the marketing analysis can be started, which will provide new strategies to innovate by following ethical and sustainable ideas.\\
  -
@@ Line 1770: / Line 1770: @@
 <WRAP centeralign>
 <figure box1>
-{{:box1.png?1000|}}
+{{:boxx1.png?1000|}}
 <caption>3D model 1</caption>
 </figure>
@@ Line 1778: / Line 1778: @@
 <WRAP center centeralign>
 <figure box2>
-{{:box2.png?1000|}} <caption>3D model 2</caption>
+{{:boxx2.png?1000|}} <caption>3D model 2</caption>
 </figure>
 </WRAP>
@@ Line 1785: / Line 1785: @@
 <WRAP center centeralign>
 <figure box3>
-{{:box3.png?1000|}} <caption>3D model 3</caption>
+{{:boxx3.png?1000|}} <caption>3D model 3</caption>
 </figure>
 </WRAP>
@@ Line 2663: / Line 2663: @@
 </figure>
 </WRAP>
+Now we are going to see how fast is the API to respond as the previous tests were more of stress tests. Table{{ref>test_table1}} holds the functional and performance (exchanged data size and latency) results of the implemented API. The
+latency tests were made by making ten consecutive calls to each operation and
+calculating the corresponding average and standard deviation values.
+Table{{ref>test_table2}} presents the load impact in the longest operation (worst case). The load
+tests consist in making 10, 100 and 1000 simultaneous requests to the operation
+with higher average latency in table{{ref>test_table1}}.
+<WRAP center centeralign>
+<table test_table1 >
+<caption>API: functional and performance results</caption>
+<WRAP box center leftalign 400px>
+^ Operation ^ Method ^ Result ^ Size ( B ) ^ Average Latency (ms) ^ Latency Deviation (ms) ^
+|Check | token | GET | OK | 207 | 6 | 0
+|Connect | POST | OK | 332 | 9 | 7 |
+|Disconnect | DELETE | OK | 334 | 7 | 3 |
+|Get | Data | GET | OK | 331 | 6 | 3
+|Get | User | GET | OK | 290 | 8 | 2
+|Login | POST | OK | 426 | 21 | 7 |
+|Logout | DELETE | OK | 245 | 25 | 10 |
+|Register | PUT | OK | 315 | 14 | 1 |
+</WRAP>
+</table>
+</WRAP>
+<WRAP center centeralign>
+<table test_table2 >
+<caption>API: load results</caption>
+<WRAP box center leftalign 400px>
+^Requests/Results ^ Operation ^ Method ^ Size ( B ) ^ Average Latency (ms) ^ Latency Deviation (ms) ^
+|10/10 | Logout | GET | 245 | 36 | 11 |
+|100/100 | Logout | GET | 245 | 28 | 8 |
+|1000/1000 | Logout | GET | 245 | 763 | 331 |
+</WRAP>
+</table>
+</WRAP>
+These results show that the app provides a friendly user experience.
  Mqtt broker testing \\
 For the mqtt broker testing the [[https://github.com/inovex/mqtt-stresser|mqtt-stresser tool from inovex]] has been used. This tool works by creating a given number of clients and a given number of messages per client. Then all clients send messages continuously and each of them report if there is any missing message. The test has been done first using 10 clients sending 150 each and the second time 100 clients sending 50 messages each. Figures {{ref>mqtt_test1}} and {{ref>mqtt_test2}} show the results.