識別跟縮減觸發網通裝置錯誤的網路流量

Abstract

在將網路設備公開到市場上之前，測試是一種能夠保證產品品質與穩定性的方法，而其中一種測試網路設備的方式是透過重播人工或是真實世界的網路流量來進行測試，相較於使用人工網路流量，透過真實流量的重播測試可以呈現出更多真實的特性。然而大量的真實網路流量及長時間的重播測試，為了減少重播時間及能有效率的重現待測產品的錯誤，如何縮減產生網路產品錯誤的流量是必須面對的問題。我們提出了透過線性搜尋與二元搜尋演算法的線性縮減與二元縮減演算法來進行流量長度的縮減。為了評估我們的方法對流量縮減的有效性，我們使用縮減比率來做為評估單位。從錯誤的分佈評估來看，在所有的錯誤中，ICMP和HTTP造成的錯誤占了所有網路設備錯誤的60%，另外，我們提出的線性縮減與二元縮減分別對於流量達到了77%與80%的縮減比率，根據結果來看，測試的時間也依照同樣的比例減少，證明我們的縮減方法可以對於網路產品測試的效率有大幅度的改善。

Testing networking devices before releasing them to the market is a way of ensuring their quality and robustness. Replay technique with artificial or real-world traffic is a method used to test networking devices. Using real-world traffic is more preferable as it displays more realistic properties in comparison with artificial traffic. The challenges with testing products with real-world traffic are mainly the high volume of the captured traces and the prolonged time required for the replay testing. In order to efficiently reproduce the defects produced by networking devices and reduce the replay time, it is necessary to reduce the size of the traces that trigger the defect of networking devices. In order to perform a defect-based traces downsizing, we propose a binary and a linear downsizing algorithms respectively based on binary and linear search algorithms. A metric called downsizing ratio is defined is order to evaluate the efficiency of the traces downsizing. The evaluation of the defects distribution show that ICMP and HTTP defects represent around 60% of the total number of defects experienced by networking devices. In addition, the binary downsizing and linear downsizing achieve a downsizing ratio of 77% and 80% respectively. As a result, the time needed to perform testing is reduced by the same amount. This is significant since the downsizing techniques used can improve the testing of networking devices.