**5.1 Parameters**

142 Grid Computing – Technology and Applications, Widespread Coverage and New Horizons

1. **Basic rule (br)** is defined as *br* (*id, rul, MRS*), where *id* is its identifier; *rul* is the description of *br*; *MRS* is the meta-rules set for revising *br*; The **basic rule set (BRS)** is

2. **Dynamic rule (dr)** is defined as dr (ct, nt, br, rul, w, sta, life), where ct CTS, nt NTS, br BRS; rul is the formalization description of dr; w is the its weight value; and sta is its state, and sta {"Naive", "Trainable", "Stable"}; "Naive" denotes that the dr is a new rule; "Trainable" denotes that the dr is revising rule; "Stable" denotes that the dr is a

3. If *dr* is a dynamic rule and *dr.w*>*MaxWeight*, which *MaxWeight* is a constant in *GMG,* we

4. If *dr* is a dynamic rule and *dr.w*<*MinWeight*, which *MinWeight* is a constant in *GMG,* we

The dynamic knowledge is the set of all the dynamic rules in *GMG*. The static knowledge is the set of all static rules. The basic knowledge can be earned by passive learning. To adapt the variety of cache resources, the dynamic rules must be generated at the start of the cache and revised during the caching process. Therefore, reinforcement learning can be adopted in the revising mechanism. Resources utilization for cache is very important reinforcement factors. Suppose that *Y1* is the *castoff threshold*, and *Y2* is the *mature threshold*; *Q (urt)* denotes as the *reinforcement function,* and *Q(urt)* >0.*urt* indicates resources utilization rate; *MaxWeight* is the maximum of the rule weight, and *MinWeight* is the minimum of the rule weight, and let *MinWeight<Y1<Y2< MaxWeight*; *MaxLife* be the maximum of life value. The revising process

*Dynamic Rule* 

*Stable*

*Trainable*

*Static Rule*

1. Suppose that a cache agent *CA* adopted a dynamic rule *dr* of *CA.KS*;

the set of all basic rules that *GMG* includes;

call *dr* as the **static rule (sr)**, its state is *"Static"*;

call *dr* as **castoff rule (cr)**. Its state is *"Castoff"*.

*Castoff Rule*

*MRS*

*Rule Naive*

mature rule; life is the its life value;

The state graph of rules is presented in figure 3.

*Basic* 

Fig. 3. The state graph of rules

2. *dr.life* + +; //increase the value of life

4. If *urt* >0 then *dr.w*=*dr.w* + *Q*(*urt*); //increase weight If *urt* <0 then *dr.w*=*dr.w-Q* (*urt*); //decrease weight 5. If dr.w>MaxWeight then dr.w=MaxWeight; If dr.w<MinWeight then dr.w=MinWeight;

3. wait for the *urt* from *MAS*;

is as follows:

Suppose that there are p cache nodes CN1, CN2 … CNp, and their memory storage ability are Ma1, Ma2 … Map, their computing ability are Ca1, Ca2 … Cap; so, the total memory storage ability is Gmc= 1 *i i p Ma* , and the total computing ability is Gcc= 1 *i i p Ca* ; and, there are m

cache files {CF1, CF2 … CFm} in PCMGM and their segment lengths are {Segl1,Segl2 … Seglm}. The parameters are as follows:

Average segment length is Asl= ( 1 *i i m Segl* )/*m*;

**Total cache ability** of Grid memory is Tc= *Gmc Asl* , and Tc denotes that the total numbers of DSEG can be stored by grid memories at one moment;

**Average cache ability** of cache node is Ac= *Tc* / *p* , and Ac denotes that the average numbers of DSEG can be stored by one cache node memory at one moment;

**Average computing ability demand of cache agent** is Ada that can be determined by the RDV of all cache nodes.

**Total cache agent ability** is Tca= *Gcc Ada* , and Tca denotes that the total numbers of cache

agent can be supported by PCMGM at one moment;

**Average cache agent ability** of cache node is Aca= / *Tca p* , and Aca denotes that the total numbers of cache agents can be supported by one cache node at one moment;

**Hot segment threshold** is a constant H in PCMGM, and if the duplication width of a segment is more than H, we call it as the hot segment. The file which includes the hot segment is called as the hot file.
