Today we continue to discuss the paper Two layered access control for Storage Area Network by Tao, DeJiao, ShiGuang. In this paper they describe it in the context of an artificial immune algorithm that can efficiently detect abnormal access. The two layers consist of a central metadata server which forms the top layer and the intelligent disks that may number 1 to n which forms the lower layer  Detectors are used in both layers that receive, analyze and inspect the access request. The inspection is the new addition to the tasks of the metadata server or intelligent disk and it intercepts the access request prior to the execution of the command and the return of the data. The immune algorithm first does a negative selection to generate detectors for access request inspection. If there is a match, the request is considered abnormal and denied. The algorithm also decides the generation and distribution of detector to intercept the access requests.  We now look at the detector generation algorithm. It uses the notion of an antigen and a detector both are represented by binary strings. The latter represents a space vector. All non-repeating binary strings are generated as the initial detector. The initial detectors that did not match any of the illegal access request were selected to be a mature detector. There are more than one detector generation algorithms namely Enumeration generation algorithm, linear generation algorithm, and greedy generation algorithm. In these algorithms, the initial detector are enumerated and the mature detectors are randomly selected.  They have large time and space overhead. For selecting mature detectors and for inspecting access requests, matching is done based on a set of matching rules. These can be r-contiguous matching rules, r-chunk matching rules and Hamming distance matching rule.  Matching involves comparing binary substrings unto r bits between the detectors and the legal access request All substrings with more than r bits in legal access request was traversed and there was no index for them. This study used the Hamming distance matching rule. The binary string matching could be improved. The number type detector and the matching threshold of r bits is defined for the length of the substring. The substring in the detector is converted to a single integer value. Access request inspection then involves analyzing legal access request and the integer value selection for number type detector.
#codingexercise
Double GetAlternateEvenNumberRangePowerRtProductpower()(Double [] A, int n, int m)
{
if (A == null) return 0;
Return A.AlternateEvenNumberRangecubePowerRtProductPower(n,m);
}