Lines Matching refs:affinity
30 =head2 Xen and NUMA machines: the concept of I<node-affinity>
33 I<node-affinity>. The node-affinity of a domain is the set of NUMA nodes
36 unrelated with the vCPU (hard and soft, see below) scheduling affinity,
41 subsystems, the domain node-affinity and the vCPUs affinity are not
43 In fact, if the domain node-affinity is not explicitly specified by the
45 basing on the vCPUs' scheduling affinity.
47 Notice that, even if the node affinity of a domain may change on-line,
56 scheduling affinity of the domain's vCPUs to the pCPUs of the node. This
75 its node-affinity. Only if that turns out to be impossible, it will just
80 Starting from Xen 4.5, credit1 supports two forms of affinity: hard and
82 soft affinity, stating where such vCPU prefers to execute on. This is
83 less strict than what it (also starting from 4.5) is called hard affinity,
87 soft affinity of the vCPUs of a domain with its node-affinity.
89 In fact, as it was for 4.3, if all the pCPUs in a vCPU's soft affinity
103 a vCPU I<is pinned> to some pCPUs and I<does not have> any soft affinity
110 a vCPU I<has> its own soft affinity and I<is not> pinned to any particular
113 affinity;
118 a vCPU I<has> its own vCPU soft affinity and I<is also> pinned to some
121 that also forms its soft affinity. In case pinning and soft affinity
122 form two disjoint sets of pCPUs, pinning "wins", and the soft affinity
145 hypervisor constructs the node-affinity of a VM basing right on its
156 to specify the soft affinity for all the vCPUs of the domain. This affects
163 if only "cpus_soft=" is present, the VM's node-affinity will be equal
164 to the nodes to which the pCPUs in the soft affinity mask belong;
169 if both "cpus_soft=" and "cpus=" are present, the VM's node-affinity
171 soft affinity belong.
181 If it finds one (some), the domain's node affinity get set to there,
185 procedure will become the soft affinity of all the vCPUs of the domain.
252 successful), it will affect the domain's node-affinity and I<not> its
266 The concept of vCPU soft affinity has been introduced for the first time
267 in Xen 4.5. In 4.3, it is the domain's node-affinity that drives the
268 NUMA-aware scheduler. The main difference is soft affinity is per-vCPU,
269 and so each vCPU can have its own mask of pCPUs, while node-affinity is
271 soft affinity.
283 all in xl or libxl, and hence no node-affinity, vCPU affinity or pinning