Common use of Scenario Clause in Contracts

Scenario. To provide a straightforward evaluation, the selected evalu- ation scenario describes the application of the approach using the response time metric, which is a QoS-related (e.g., latency, throughput, and packet loss) SLA. The metrics in QoS- related SLAs are quantitative, being able to be automatically verified in the SC. Nevertheless, the designed approach can be modified to fit different types of SLAs, such as service availability. The evaluation scenario was composed of a web server and a client performing periodic requests. A Raspberry Pi model B with a quad-core ARM Cortex-A7 CPU @ 900 MHz, 1 GB of RAM, and a 150 Mbps Wi-Fi dongle hosted an Apache web server, and a Lenovo ThinkPad T430 with a quad-core Intel(R) Core(TM) i7-3520M CPU @ 2.90 GHz, 16 GB of RAM, and a 1 Gbps Ethernet connection performed the requests as the client. The client requested at every 1 s a PHP page that performed random calculations to introduce load in the server. Moreover, after each finished request, the client retrieved the response time of the request and stored this value in a Comma- Separated-Value (CSV) file. During a 30 minutes interval, 1800 requests were performed and measured (30 min 60 s = 1800); the response times of these requests are depicted in Figure 2. It can be seen in the graph that the majority of the requests were answered in 0.2 s to 0.3 s, and that there are requests with response times over 0.6 s. The average response time for the period was 0.2445 s. Based on these values, the violation threshold was fixed to 0.3 s, which is depicted using a red dashed line in the graph. 0.9 0.8 Response Time (seconds) 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Response Time Violation Threshold Fig. 2: Response Time between the Client and the Web Server Listing 6: Compensate and Deposit Functions