Each processor in an MP system resides within a processor cluster. At reset time (i.e., the trailing edge of the reset signal), each processor samples two signal groups to determine what cluster it resides in, as well as what its unique ID is.
The Cluster ID identifies what cluster of processors the processor (and therefore its encapsulated Local APIC) is a member of. The processor may be assigned a cluster number of 0, 1, 2, or 3. A[12:11]# are sampled on the trailing-edge of reset to determine the cluster ID (see Figure 61-11 on page 1513):
Sampled electrical value 00b = cluster 3.
Sampled electrical value 01b = cluster 2.
Sampled electrical value 10b = cluster 1.
Sampled electrical value ...