Bad NUMA CPU numbering on server with more than 256 CPUs

Solution Verified - Updated -

Issue

Some customers using servers with more than 256 CPUs notice an unusual numbering of the CPUs. The CPU number 0, which should normally be assigned to node 0, is assigned to a different node (node 4), for example :

[root@ls3014 ~]# numactl -H
available: 8 nodes (0-7)
node 0 cpus: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
node 0 size: 393216 MB
node 0 free: 385220 MB
node 1 cpus: 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
node 1 size: 393216 MB
node 1 free: 385299 MB
node 2 cpus: 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
node 2 size: 393216 MB
node 2 free: 385352 MB
node 3 cpus: 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
node 3 size: 393216 MB
node 3 free: 385241 MB
node 4 cpus: 0 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 216 217 218 219 220 221 222 223 224 225 226 227 228 229 

Environment

  • Red Hat Enterprise Linux (RHEL) 7.1
  • NUMA
  • Server with more than 256 CPUs

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In