An Application-Level QoS Control Method Based on Local Bandwidth Scheduling

Algorithm 1

Bandwidth dynamic scheduling method (the first allocation).

Input: The tokens generated by generator (L_token)

Output: The first token allocation policy (L_allocation)

(1) FirstAllocation (L_token)

(2) Obtain the running Web process list L_process

(3) Query the priorities of processes in L_process in the

(4) priority database

(5) Sort L_process in descending order of priority

(6) Initialize the first allocation policy L_allocation

(7) FOR i=0 to COUNT(L_process)-1  

(8) Mark the bucket of process in L_process as B_i

(9) IF ((B_i) < (B_i))

(10) // Non-full bucket

(11) Get the bandwidth V_i for the corresponding

(12) process in the priority database

(13) IF (V_i != null) // Process exists in the priority database

(14) // T_i is the remaining capacity of B_i

(15) T_i = (B_i) - (B_i)

(16) Initialize the allocation policy (allocation) for B_i

(17) // Set the bucket member of allocation (token ID)

(18) allocation.BUCKET = B_i

(19) // Set the TOKEN member of allocation (token count)

(20) // Add the smaller one between 1.2V_i and T_i to the

(21) // current token bucket

(22) allocation.TOKEN = allocation.TOKEN+MIN(T_i, 1.2V_i)

(23) Add allocation into the first allocation policy L_allocation

(24) Delete the allocated tokens from L_token

(25)

(26) ELSE // All processes in priority database checked

(27) // A null value of V_i means that the corresponding process

(28) // of B_i does not exist in the priority database and the

(29) // remaining processes do not exist in the database

(30) // either.These processes won’t be allocated any tokens

(31) // in the first allocation.

(32) RETURN L_allocation

(33)

(34)

(35)

(36) RETURN L_allocation // All processes checked

(37)