To verify the correctness of hyper rule cache, two rule sets and associated trace files are used. Each packet in the trace file is associated with the correct rule that matches it. These files can be found from the following website: http://www.arl.wustl.edu/ hs1/project/.
We randomly generate 2, 4 and 10 actions for each rule in the two rule sets, and run hyper rule cache on these 6 pairs of rule set and traces. The action of each packet that is hit in the rule cache is printed out, and is compared with its real action.
Our experiments show that for each packet hit in rule cache, hyper rule cache give the same action as its real action. The results are summarized in table 3. As can be seen in the table, in all the 6 cases, 100% of packets are correctly classified by hyper rule cache, which proves the correctness of hyper rule cache scheme.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment