Difference between revisions of "Plugin:IntelRDT"

From collectd Wiki
Jump to: navigation, search
(Created page with "{{Infobox Plugin | Name=Intel RDT | Type=read | Callbacks={{Callback|config}}, {{Callback|init}}, {{Callback|read}}, {{Callback|shutdown}} | Status={{supported}} | F...")
 
 
(9 intermediate revisions by 2 users not shown)
Line 5: Line 5:
 
   | Status={{supported}}
 
   | Status={{supported}}
 
   | FirstVersion={{Version|5.7}}
 
   | FirstVersion={{Version|5.7}}
−
   | Copyright=''2016–2018'' Intel Corporation
+
   | Copyright=''2016–2018'' Intel Corporation<br /> ''Serhiy Pshyk'' <br /> ''Mateusz Starzyk'' <br /> ''Wojciech Andralojc'' <br /> ''Michał Aleksiński''
 
   | License={{MIT_License}}
 
   | License={{MIT_License}}
 
   | Manpage={{Manpage|collectd.conf|5|plugin_intel_rdt}}
 
   | Manpage={{Manpage|collectd.conf|5|plugin_intel_rdt}}
 
}}
 
}}
−
The ''intel_rdt'' plugin collects information provided by monitoring features of
+
The ''intel_rdt'' plugin collects information provided by monitoring features of Intel Resource Director Technology (Intel(R) RDT) like Cache Monitoring Technology (CMT), Memory Bandwidth Monitoring (MBM), Cache Allocation Technology (CAT) and Code and Data Prioritization (CDP) Technology provide the hardware framework to monitor and control the utilization of shared resources, like last level cache, memory bandwidth. These Technologies comprise Intel’s Resource Director Technology (RDT). As multithreaded and multicore platform architectures emerge, running workloads in single-threaded, multithreaded, or complex virtual machine environment, the last level cache and memory bandwidth are key resources to manage. Intel introduces CMT, MBM, CAT and CDP to manage these various workloads across shared resources. 
−
Intel Resource Director Technology (Intel(R) RDT) like Cache Monitoring
+
 
−
Technology (CMT), Memory Bandwidth Monitoring (MBM). These features provide
+
{| class="wikitable" style="background-color:#FFF;"
−
information about utilization of shared resources.
+
|-
 +
! <br />Name
 +
! <br />Type
 +
! <br />Type Instance
 +
! <br />Description
 +
! <br />Comment
 +
|-
 +
| <br />LLC
 +
| <br />bytes
 +
| <br />llc
 +
| <br />last  level cache occupancy (CMT)
 +
| <br />Existing  type
 +
|-
 +
| <br />MBL
 +
| <br />memory_bandwidth
 +
| <br />local
 +
| <br />the  bandwidth of accessing memory associated with the local socket (MBM)
 +
| <br />Existing  type
 +
|-
 +
| <br />MBR
 +
| <br />memory_bandwidth
 +
| <br />remote
 +
| <br />the  bandwidth of accessing the remote socket (MBM)
 +
| <br />Existing  type
 +
|-
 +
| <br />IPC
 +
| <br />ipc
 +
| <br />
 +
| <br />instructions  per clock
 +
| <br />New  type introduced in types.db
 +
|}
 +
 
 
For a full description of available options please refer to the {{Manpage|collectd.conf|5|plugin_intel_rdt}} manual page.
 
For a full description of available options please refer to the {{Manpage|collectd.conf|5|plugin_intel_rdt}} manual page.
  
Line 21: Line 52:
 
  </Plugin>
 
  </Plugin>
  
−
== Example graphs ==
+
=== Parameters ===
 +
 
 +
{| class="wikitable"
 +
|-
 +
! <br />Name
 +
! <br />Description
 +
! <br />Comment
 +
|-
 +
| <br />Interval
 +
| <br />The interval  within which to retrieve statistics on monitored events in seconds
 +
| <br />Interval option is supported by collectd  and is defined in <LoadPlugin> block. No additional functionality  should be developed in intel_rdt plugin  to support this option.
 +
|-
 +
| <br />Cores
 +
| <br />Core groups  definition. Monitored metrics are reported as aggregated statistics per  group.
 +
| <br />The field is  represented as list of strings with core group values. Each string represents  a list of cores in a group. Allowed formats are: “0,1,2,3” “0-10,20-18”  “1,3,5-8,10,0x10-12”.<br />  <br />If an empty string is provided as value for this field default cores  configuration should be applied - a separate group for each core.
 +
|}
 +
 
 +
== Metrics ==
 +
 
 +
{| class="wikitable" style="background-color:#FFF;"
 +
|-
 +
! <br />Metric/Feature/Input
 +
! <br />Name
 +
! <br />Date Type
 +
! <br />Format Example
 +
! <br />Description
 +
! <br />Dependencies
 +
! <br />Limitations
 +
! <br />Comments
 +
|-
 +
| <br />Metric
 +
| <br />Memory  Bandwidth on Local NUMA Node
 +
| <br />Bytes/Second
 +
| <br />3934325
 +
| <br />Memory  bandwidth utilization by the relevant CPU core on the local NUMA memory  channel
 +
| <br />PQOS  ToolSet
 +
| <br />Does  not provide the value per process basis due to lack of resctrl fs support
 +
| <br />Dependent  on PQOS toolset to read the metric value
 +
|-
 +
| <br />Metric
 +
| <br />Memory  Bandwidth on Remote NUMA Node
 +
| <br />Bytes/Second
 +
| <br />3934325
 +
| <br />Memory  bandwidth utilization by the relevant CPU core on the remote NUMA memory  channel
 +
| <br />PQOS  ToolSet
 +
| <br />Does  not provide the value per process basis due to lack of resctrl fs support
 +
| <br />Dependent  on PQOS toolset to read the metric value
 +
|-
 +
| <br />Metric
 +
| <br />Total  Memory Bandwidth
 +
| <br />Bytes/Second
 +
| <br />3934325
 +
| <br />Total  memory bandwidth utilized by a CPU core on local and remote NUMA memory  channels
 +
| <br />PQOS  ToolSet
 +
| <br />Does  not provide the value per process basis due to lack of resctrl fs support
 +
| <br />Not  part of the Collectd plugin
 +
|-
 +
| <br />Metric
 +
| <br />L3  Cache Occupancy
 +
| <br />Bytes
 +
| <br />45345434
 +
| <br />Total  Last Level Cache occupancy by a CPU core
 +
| <br />PQOS  ToolSet
 +
| <br />Does  not provide the value per process basis due to lack of resctrl fs support
 +
| <br />Dependent  on PQOS toolset to read the metric value
 +
|-
 +
| <br />Metric
 +
| <br />Instructions  Per Cycle
 +
| <br />Integer
 +
| <br />23734
 +
| <br />Total  instructions per cycle executed by a CPU core
 +
| <br />PQOS  ToolSet
 +
| <br />None
 +
| <br />Dependent  on PQOS toolset to read the metric value
 +
|-
 +
| <br />Input
 +
| <br />Cores
 +
| <br />Integer  Array
 +
| <br />[0-12]  or [1,2,3]
 +
| <br />The  list of CPU core(s) to be provided as input by the user for which the  corresponding metrics are required
 +
| <br />None
 +
| <br />None
 +
| <br />Configuration  input in the plugin .conf file
 +
|-
 +
| <br />Input
 +
| <br />Configuration  Interval
 +
| <br />Integer
 +
| <br />1  or 10
 +
| <br />The  interval in seconds at which the metrics need to be collectd
 +
| <br />None
 +
| <br />None
 +
| <br />Configuration  input in the plugin .conf file
 +
|-
 +
| <br />Metric
 +
| <br />Memory  Bandwidth of the process on Local NUMA Node
 +
| <br />Bytes
 +
| <br />3934325
 +
| <br />Memory  bandwidth utilization by the relevant process on the local NUMA memory  channel
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|-
 +
| <br />Metric
 +
| <br />Memory  Bandwidth of the process on Remote NUMA Node
 +
| <br />Bytes
 +
| <br />3934325
 +
| <br />Memory  bandwidth utilization by the relevant process on the remote NUMA memory  channel
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|-
 +
| <br />Metric
 +
| <br />Total  Memory Bandwidth of the process
 +
| <br />Bytes
 +
| <br />3934325
 +
| <br />Total  memory bandwidth utilized by a process on local and remote NUMA memory  channels
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|-
 +
| <br />Metric
 +
| <br />L3  Cache Occupancy of the process
 +
| <br />Bytes
 +
| <br />45345434
 +
| <br />Total  Last Level Cache occupancy by a process
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|-
 +
| <br />Metric
 +
| <br />Instructions  Per Cycle of the process
 +
| <br />Integer
 +
| <br />23734
 +
| <br />Total  instructions per cycle executed by a process
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|-
 +
| <br />Input
 +
| <br />Process  List
 +
| <br />Array
 +
| <br />[qemu,  pmd]
 +
| <br />List  of processes for which metrics are to be collected
 +
| <br />resctrl  fs & kernel 4.14
 +
| <br />TBD
 +
| <br />Still  to create Collectd PR
 +
|}
 +
 
 +
== Example graph ==
  
−
[TODO]
+
[[File:rdt_llc.png]]
  
 
== Dependencies ==
 
== Dependencies ==
  
−
* [http://https://github.com/intel/intel-cmt-cat intel-cmt-cat]
+
* [https://github.com/intel/intel-cmt-cat intel-cmt-cat]
 +
 
 +
== See also ==
 +
 
 +
* [[Plugin:IntelRDT/tests]]
 +
* [https://wiki.opnfv.org/display/fastpath/Intel_RDT Intel RDT plugin high level design document]
  
 
[[Category:Plugins]]
 
[[Category:Plugins]]
 +
[[Category:Needs Info]]
 
{{DEFAULTSORT:Intel_Rdt}}
 
{{DEFAULTSORT:Intel_Rdt}}

Latest revision as of 16:19, 16 June 2020

Intel RDT plugin
Type: read
Callbacks: config, init, read, shutdown
Status: supported
First version: 5.7
Copyright: 2016–2018 Intel Corporation
Serhiy Pshyk
Mateusz Starzyk
Wojciech Andralojc
Michał Aleksiński
License: MIT license
Manpage: collectd.conf(5)
List of Plugins

The intel_rdt plugin collects information provided by monitoring features of Intel Resource Director Technology (Intel(R) RDT) like Cache Monitoring Technology (CMT), Memory Bandwidth Monitoring (MBM), Cache Allocation Technology (CAT) and Code and Data Prioritization (CDP) Technology provide the hardware framework to monitor and control the utilization of shared resources, like last level cache, memory bandwidth. These Technologies comprise Intel’s Resource Director Technology (RDT). As multithreaded and multicore platform architectures emerge, running workloads in single-threaded, multithreaded, or complex virtual machine environment, the last level cache and memory bandwidth are key resources to manage. Intel introduces CMT, MBM, CAT and CDP to manage these various workloads across shared resources. 


Name
Type
Type Instance
Description
Comment

LLC
bytes
llc
last level cache occupancy (CMT)
Existing type

MBL
memory_bandwidth
local
the bandwidth of accessing memory associated with the local socket (MBM)
Existing type

MBR
memory_bandwidth
remote
the bandwidth of accessing the remote socket (MBM)
Existing type

IPC
ipc

instructions per clock
New type introduced in types.db

For a full description of available options please refer to the collectd.conf(5) manual page.

Synopsis

<Plugin "intel_rdt">
  Cores "0-2" "3,4,6" "8-10,15"
</Plugin>

Parameters


Name
Description
Comment

Interval
The interval within which to retrieve statistics on monitored events in seconds
Interval option is supported by collectd and is defined in <LoadPlugin> block. No additional functionality should be developed in intel_rdt plugin to support this option.

Cores
Core groups definition. Monitored metrics are reported as aggregated statistics per group.
The field is represented as list of strings with core group values. Each string represents a list of cores in a group. Allowed formats are: “0,1,2,3” “0-10,20-18” “1,3,5-8,10,0x10-12”.

If an empty string is provided as value for this field default cores configuration should be applied - a separate group for each core.

Metrics


Metric/Feature/Input
Name
Date Type
Format Example
Description
Dependencies
Limitations
Comments

Metric
Memory Bandwidth on Local NUMA Node
Bytes/Second
3934325
Memory bandwidth utilization by the relevant CPU core on the local NUMA memory channel
PQOS ToolSet
Does not provide the value per process basis due to lack of resctrl fs support
Dependent on PQOS toolset to read the metric value

Metric
Memory Bandwidth on Remote NUMA Node
Bytes/Second
3934325
Memory bandwidth utilization by the relevant CPU core on the remote NUMA memory channel
PQOS ToolSet
Does not provide the value per process basis due to lack of resctrl fs support
Dependent on PQOS toolset to read the metric value

Metric
Total Memory Bandwidth
Bytes/Second
3934325
Total memory bandwidth utilized by a CPU core on local and remote NUMA memory channels
PQOS ToolSet
Does not provide the value per process basis due to lack of resctrl fs support
Not part of the Collectd plugin

Metric
L3 Cache Occupancy
Bytes
45345434
Total Last Level Cache occupancy by a CPU core
PQOS ToolSet
Does not provide the value per process basis due to lack of resctrl fs support
Dependent on PQOS toolset to read the metric value

Metric
Instructions Per Cycle
Integer
23734
Total instructions per cycle executed by a CPU core
PQOS ToolSet
None
Dependent on PQOS toolset to read the metric value

Input
Cores
Integer Array
[0-12] or [1,2,3]
The list of CPU core(s) to be provided as input by the user for which the corresponding metrics are required
None
None
Configuration input in the plugin .conf file

Input
Configuration Interval
Integer
1 or 10
The interval in seconds at which the metrics need to be collectd
None
None
Configuration input in the plugin .conf file

Metric
Memory Bandwidth of the process on Local NUMA Node
Bytes
3934325
Memory bandwidth utilization by the relevant process on the local NUMA memory channel
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Metric
Memory Bandwidth of the process on Remote NUMA Node
Bytes
3934325
Memory bandwidth utilization by the relevant process on the remote NUMA memory channel
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Metric
Total Memory Bandwidth of the process
Bytes
3934325
Total memory bandwidth utilized by a process on local and remote NUMA memory channels
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Metric
L3 Cache Occupancy of the process
Bytes
45345434
Total Last Level Cache occupancy by a process
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Metric
Instructions Per Cycle of the process
Integer
23734
Total instructions per cycle executed by a process
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Input
Process List
Array
[qemu, pmd]
List of processes for which metrics are to be collected
resctrl fs & kernel 4.14
TBD
Still to create Collectd PR

Example graph

Rdt llc.png

Dependencies

See also

Retrieved from "https://collectd.org/wiki/index.php?title=Plugin:IntelRDT&oldid=5020"