[pve-devel] [PATCH pve-ha-manager 0/8] WIP: ressource aware HA manager + balancer

Alexandre Derumier aderumier at odiso.com
Wed Apr 27 17:33:43 CEST 2022


Hi,
This is a big rework of my previous patches series.
Currently, work in progress, so you don't need to review it for now.
(but if you want, comments are still welcome ;)

Biggest changes:

 - This now use a new ranking algorithm: AHP-Topsis. (details are in patch)
 - Adding a balancer. The main idea is to do something like drs 2.0 on vmware 7.
   The balancer is vm centric, looking at health of the vms 
   (mainly based on the pressure). (more details in the patch too)
 - add affifinity/anti affinity vm ha groups.


I'm currently working on adding a lot of tests, but current tests are already
working fine.
It's also need missing vm pressure stats, ksm stats and some other counters.

I have tried to split code in differents patch as much as possible

Alexandre

Alexandre Derumier (8):
  add AHP && Topsis Math Helpers
  get services && nodes stats
  implement ressource aware service recovery
  add vm loadbalancing
  sim : hardware: add read stats
  add resources group
  add affinity
  add tests

 debian/pve-ha-manager.install                 |   6 +
 src/PVE/HA/Balancer/AHP.pm                    | 120 ++++++++++
 src/PVE/HA/Balancer/Makefile                  |   6 +
 src/PVE/HA/Balancer/Nodes.pm                  | 219 ++++++++++++++++++
 src/PVE/HA/Balancer/Services.pm               | 203 ++++++++++++++++
 src/PVE/HA/Balancer/Stats.pm                  | 134 +++++++++++
 src/PVE/HA/Balancer/Topsis.pm                 | 115 +++++++++
 src/PVE/HA/Config.pm                          |  22 ++
 src/PVE/HA/Env.pm                             |  45 ++++
 src/PVE/HA/Env/PVE2.pm                        | 207 +++++++++++++++++
 src/PVE/HA/Makefile                           |   4 +-
 src/PVE/HA/Manager.pm                         | 101 +++++++-
 src/PVE/HA/ResourcesGroups.pm                 |  90 +++++++
 src/PVE/HA/Sim/Env.pm                         |   8 +
 src/PVE/HA/Sim/Hardware.pm                    | 165 +++++++++++++
 src/PVE/HA/Sim/TestEnv.pm                     |  68 +++++-
 src/PVE/HA/Tools.pm                           |   6 +
 src/test/test-basic0-balance-affinity/README  |   1 +
 src/test/test-basic0-balance-affinity/cmdlist |   3 +
 .../datacenter.cfg                            |   5 +
 .../hardware_status                           |   5 +
 .../test-basic0-balance-affinity/log.expect   |  40 ++++
 .../manager_status                            |   1 +
 .../test-basic0-balance-affinity/node_stats   |   5 +
 .../resources_groups                          |   3 +
 .../service_config                            |   5 +
 .../service_stats                             |   5 +
 .../test-basic0-balance-cpupressure/README    |   1 +
 .../test-basic0-balance-cpupressure/cmdlist   |   3 +
 .../datacenter.cfg                            |   5 +
 .../hardware_status                           |   5 +
 .../log.expect                                |  40 ++++
 .../manager_status                            |   1 +
 .../node_stats                                |   5 +
 .../service_config                            |   5 +
 .../service_stats                             |   5 +
 src/test/test-basic0-recovery-affinity/README |   1 +
 .../test-basic0-recovery-affinity/cmdlist     |   4 +
 .../datacenter.cfg                            |   5 +
 .../hardware_status                           |   5 +
 .../test-basic0-recovery-affinity/log.expect  |  53 +++++
 .../manager_status                            |   1 +
 .../test-basic0-recovery-affinity/node_stats  |   5 +
 .../resources_groups                          |   3 +
 .../service_config                            |   5 +
 .../service_stats                             |   5 +
 .../test-basic0-recovery-antifinity/README    |   1 +
 .../test-basic0-recovery-antifinity/cmdlist   |   4 +
 .../datacenter.cfg                            |   5 +
 .../hardware_status                           |   5 +
 .../log.expect                                |  53 +++++
 .../manager_status                            |   1 +
 .../node_stats                                |   5 +
 .../resources_groups                          |   3 +
 .../service_config                            |   5 +
 .../service_stats                             |   5 +
 src/test/test-basic00-recovery-storage/README |   1 +
 .../test-basic00-recovery-storage/cmdlist     |   4 +
 .../hardware_status                           |   5 +
 .../test-basic00-recovery-storage/log.expect  |  52 +++++
 .../manager_status                            |   1 +
 .../service_config                            |   5 +
 .../test-basic00-recovery-storage/storecfg    |   3 +
 .../test-basic00-recovery-storage/vm_config   |   5 +
 64 files changed, 1902 insertions(+), 10 deletions(-)
 create mode 100644 src/PVE/HA/Balancer/AHP.pm
 create mode 100644 src/PVE/HA/Balancer/Makefile
 create mode 100644 src/PVE/HA/Balancer/Nodes.pm
 create mode 100644 src/PVE/HA/Balancer/Services.pm
 create mode 100644 src/PVE/HA/Balancer/Stats.pm
 create mode 100644 src/PVE/HA/Balancer/Topsis.pm
 create mode 100644 src/PVE/HA/ResourcesGroups.pm
 create mode 100644 src/test/test-basic0-balance-affinity/README
 create mode 100644 src/test/test-basic0-balance-affinity/cmdlist
 create mode 100644 src/test/test-basic0-balance-affinity/datacenter.cfg
 create mode 100644 src/test/test-basic0-balance-affinity/hardware_status
 create mode 100644 src/test/test-basic0-balance-affinity/log.expect
 create mode 100644 src/test/test-basic0-balance-affinity/manager_status
 create mode 100644 src/test/test-basic0-balance-affinity/node_stats
 create mode 100644 src/test/test-basic0-balance-affinity/resources_groups
 create mode 100644 src/test/test-basic0-balance-affinity/service_config
 create mode 100644 src/test/test-basic0-balance-affinity/service_stats
 create mode 100644 src/test/test-basic0-balance-cpupressure/README
 create mode 100644 src/test/test-basic0-balance-cpupressure/cmdlist
 create mode 100644 src/test/test-basic0-balance-cpupressure/datacenter.cfg
 create mode 100644 src/test/test-basic0-balance-cpupressure/hardware_status
 create mode 100644 src/test/test-basic0-balance-cpupressure/log.expect
 create mode 100644 src/test/test-basic0-balance-cpupressure/manager_status
 create mode 100644 src/test/test-basic0-balance-cpupressure/node_stats
 create mode 100644 src/test/test-basic0-balance-cpupressure/service_config
 create mode 100644 src/test/test-basic0-balance-cpupressure/service_stats
 create mode 100644 src/test/test-basic0-recovery-affinity/README
 create mode 100644 src/test/test-basic0-recovery-affinity/cmdlist
 create mode 100644 src/test/test-basic0-recovery-affinity/datacenter.cfg
 create mode 100644 src/test/test-basic0-recovery-affinity/hardware_status
 create mode 100644 src/test/test-basic0-recovery-affinity/log.expect
 create mode 100644 src/test/test-basic0-recovery-affinity/manager_status
 create mode 100644 src/test/test-basic0-recovery-affinity/node_stats
 create mode 100644 src/test/test-basic0-recovery-affinity/resources_groups
 create mode 100644 src/test/test-basic0-recovery-affinity/service_config
 create mode 100644 src/test/test-basic0-recovery-affinity/service_stats
 create mode 100644 src/test/test-basic0-recovery-antifinity/README
 create mode 100644 src/test/test-basic0-recovery-antifinity/cmdlist
 create mode 100644 src/test/test-basic0-recovery-antifinity/datacenter.cfg
 create mode 100644 src/test/test-basic0-recovery-antifinity/hardware_status
 create mode 100644 src/test/test-basic0-recovery-antifinity/log.expect
 create mode 100644 src/test/test-basic0-recovery-antifinity/manager_status
 create mode 100644 src/test/test-basic0-recovery-antifinity/node_stats
 create mode 100644 src/test/test-basic0-recovery-antifinity/resources_groups
 create mode 100644 src/test/test-basic0-recovery-antifinity/service_config
 create mode 100644 src/test/test-basic0-recovery-antifinity/service_stats
 create mode 100644 src/test/test-basic00-recovery-storage/README
 create mode 100644 src/test/test-basic00-recovery-storage/cmdlist
 create mode 100644 src/test/test-basic00-recovery-storage/hardware_status
 create mode 100644 src/test/test-basic00-recovery-storage/log.expect
 create mode 100644 src/test/test-basic00-recovery-storage/manager_status
 create mode 100644 src/test/test-basic00-recovery-storage/service_config
 create mode 100644 src/test/test-basic00-recovery-storage/storecfg
 create mode 100644 src/test/test-basic00-recovery-storage/vm_config

-- 
2.30.2





More information about the pve-devel mailing list