Know the Hardware

From RCSWiki
Revision as of 00:07, 10 January 2020 by Darcy (talk | contribs)
Jump to navigation Jump to search

All About the Hardware

A cluster is a collection of independent computers referred to as nodes that are tightly coupled together to provide scalable computing resources. It will have a central control node that you access to setup and start your jobs. The hardware you are using on a cluster system can be based on various types of processors from many companies. Most common are Intel based systems, AMD, IBM, or ARM. They may include GPU (Graphics Processing Unit) chips which provide vector calculation abilities. Each node will have some amount of memory (preferably a lot) local to it. Nodes may be grouped together into partitions that reflect the type of processor, presence of a GPU, amount of memory and other hardware/software specific attributes.

The nodes within the cluster will be interconnected with some form of high speed networking. This can be leveraged to create jobs that take advantage of a technique called message passing. Depending on the type of job you have, this technique could increase the speed of your calculations.

The amount of memory available on nodes can influence where you would like a job to run.

Useful Linux Hardware Information

Linux provides a great deal of information about the hardware. To discover information about and the number of CPUs, examine the the contents of /proc/cpuinfo:

[mycomputer]$ less /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz
stepping        : 3
microcode       : 0x27
cpu MHz         : 3626.178
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 0
cpu cores       : 4
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds
bogomips        : 7183.62
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:
...

Much of this information is not important unless you are writing specific machine optimized code. An important thing to note is the number of cpu cores may be different than the number of processors listed. The number of cores is the important number.

To discover the amount of memory that a Linux machine has, you can use the free command:

[mycomputer]$ free -h
              total        used        free      shared  buff/cache   available
Mem:          7.7Gi       5.9Gi       293Mi       106Mi       1.5Gi       1.4Gi
Swap:         7.9Gi       4.9Gi       2.9Gi