energy efficient resource management in virtualised datacenters
DESCRIPTION
A coarse grain overview of energy-aware resource management approaches since the last ten years through 2 EU founded projects. Presented during the 2014 "Energy Aware Network" Labex day at Inria. (see http://www.ucnlab.eu/fr/node/66)TRANSCRIPT
![Page 1: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/1.jpg)
energy efficient resource management in virtualised datacenters
Fabien Hermenier
![Page 2: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/2.jpg)
(Environmental Protection Agency, 2007)
of the 2005 budget1.5%
in 2010 ?3 %USA
![Page 3: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/3.jpg)
1Consume less2003
![Page 4: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/4.jpg)
●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●
●
●●●
●
●●●●●●
●
●●●●●●●●●●●●●●●●
●●●
●●●●●●●●●●●●●●●●●●●●●●●●●
●●●●
●
●●●●●●●●●●●●●●●●●●●●●●●●
●
●●●
●●●●●●●●●●●●
●
●●●●●●●●●●●●●
●●
●●●●●●●●
●●●
●
●●●●●●●●●●●●
●●●
●●●●●●●●●●
●●●
●●●●●●●
●●●
●
●●
●
●●●●●●●●
●
●
●●●●●●●●●
●
●
●●●●●●●●●●●●
●
●
●●●●●
●
●
●●●●●●●●●●●●●●
●
●
●
●●
●●●●●●●●●●●●●●●●●
●●
●●
●●●●●
●
●●●●●●●
●●
●●
●
●
●
●
●
●
●●
●
●●●●
●●●
●●●●●
●●
●●●
●
●
●●●●●●●●●●●●●●
●
●●
●
●●●●●●●●
●
●●●●●●●●●●●●
●
●
●●
●
●
●●●●●●●●●●
●
●
●●●●
●
●
●
●●●●●●●●●●●●●●●●
●
●
●
●
●●●
●●●●
●●
●●●●●●●●●●●●●
●●
●●●●●●
●●●●
●
●
●●
●●●●●●
●●
●●●●●●●●●
●●
●●
●
●
●●●
●
●
●
●●●
●
●●
●
●●
●
●●●●●●●●
●
●
●
●●
●
●●●●●●●●
●●
●●●●●●●●●
●
●●●
●
●
●
●
●
●●
●●●●●
●
●●●
●
110
120
130
140
150
160
170
180
0 1 2 3 4VM #
Con
sum
ptio
n (W
att)
Node consumption statistics (Cluster edel, 128GB HDD)
the brown constant
110 W
175 W
![Page 5: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/5.jpg)
2003“let’s turn off useless stuff”elnozahy et al. :
Packingconcentrate WWW
requests
![Page 6: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/6.jpg)
2006 “look! look it’s moving”Hansen et al. :
live migrationN1 N2VM1
![Page 7: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/7.jpg)
1) model power consumption 2) pack VMs 3) turn-off unused nodes 4) minimize(W) 5) profit
consume less, the theory
![Page 8: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/8.jpg)
hw. heterogeneity env. capabilities performance vs. energy workload volatility data center size
,,consume less, the practice
![Page 9: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/9.jpg)
BtrPlace proposalcore reconfiguration algorithm
users scripts
+ =specialized reconfiguration algorithm
![Page 10: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/10.jpg)
spread(VM[2..3]); preserve({VM1},’ucpu’, 3); offline(@N4);
0’00 to 0’02: relocate(VM2,N2) 0’00 to 0’04: relocate(VM6,N2) 0’02 to 0’05: relocate(VM4,N1) 0’04 to 0’08: shutdown(N4) 0’05 to 0’06: allocate(VM1,‘cpu’,3)
The reconfiguration plan
BtrPlace
![Page 11: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/11.jpg)
the core reconfiguration algorithm is modeled wrt. the impact of actions on resources
![Page 12: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/12.jpg)
allDifferent(dhost
1 , d
host
2 ) ^d
host
1 = c
host
2 ! d
st
1 � c
ed
2 ^d
host
2 = c
host
1 ! d
st
2 � c
ed
1
spread({VM1,VM2}):
Premade variables, constraints
![Page 13: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/13.jpg)
Constraint Programming !
!
!
to the rescue
![Page 14: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/14.jpg)
2010 2013
Energy aware ICT optimization policies (+ btrPlace)
![Page 15: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/15.jpg)
multi-core CPUs, DDR3 memory, spinning HD, PUE / CUE, boot/shutdown time
VM template migration duration migration payback time
hw. particularities
a fine-grain power model
workload particularities
![Page 16: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/16.jpg)
energy-related variables are linked to core ones
![Page 17: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/17.jpg)
energy-oriented constraintsMaxServerPower
DelayBtwMigrationsDelayBtwServerSwitch
PayBackTime
SpareCPUs
minEnergyCons
minGasEmissions
cap consumption
reduce ping-pong effects
migration as an investment
control the consolidation aggressiveness
optimisation criteria
![Page 18: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/18.jpg)
500 1000 1500 2000 2500
1
234
5
6
7
Servers
Tim
e (m
inut
es)
P4G P4G + spare P4G + spare + vcpu P4G + spare + vcpu + delay
the core problem dominates scalability
![Page 19: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/19.jpg)
coarse to fine grain optimisation
-16%
-47%
- 27%-7%
![Page 20: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/20.jpg)
Consuming lessdealing with
THE BROWN CONSTANT
![Page 21: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/21.jpg)
Consuming lessdealing with
UNAPPROPRIATE
HARDWARE
![Page 22: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/22.jpg)
hw. community did not
chill
energy proportional servers, free cooling,
fanless processors
![Page 23: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/23.jpg)
?workload agnostic
need to revamp software approaches migration
a balancing problem
![Page 24: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/24.jpg)
2Consume better2012
![Page 25: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/25.jpg)
2013 2016
let existing and new data centres become energy adaptive
DC4Cities
![Page 26: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/26.jpg)
align workload to renewable energies availability
![Page 27: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/27.jpg)
DC4Cities gray box approach
forecasts1) 24h power budget
2) alt. working modes
3) selected mode
energy adaptive applications
![Page 28: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/28.jpg)
Energy adaptive Batch scheduling
How many parallel jobs, which ones to defer
Trade replicas against latency
Allocate resources against SLAs
Energy adaptive WWW
Energy adaptive IaaS
Energy adaptive …
…
![Page 29: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/29.jpg)
ConclusionS
![Page 30: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/30.jpg)
consume less, consume better, non-renewable power sources, DVFS, live-migration, deferrable workload, so many facets, non-deferrable workload, elasticity, VM packing, VOVO, VM balancing, white-box approach, black-box approach, solutions need to follow new capabilities and usage, priority over SLA, priority over savings, renewable power sources, fine grain power model, coarse grain power model, steady workload, bursty workload
![Page 31: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/31.jpg)
in 20101.1 - 1.5 %
(Environmental Protection Agency, 2013)
USA
![Page 32: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/32.jpg)
.org
![Page 33: energy efficient resource management in virtualised datacenters](https://reader033.vdocuments.us/reader033/viewer/2022042816/559422841a28ab21578b4579/html5/thumbnails/33.jpg)