SLIDE 1
- Introduction
- Energy Consumption
- Elastic Energy Consumption Model
- Methodology
- Results Analysis
- Conclusion
ANALYZING ENERGY CONSUMPTION OF ELASTIC HPC APPLICATIONS IN THE - - PowerPoint PPT Presentation
ANALYZING ENERGY CONSUMPTION OF ELASTIC HPC APPLICATIONS IN THE - - PowerPoint PPT Presentation
ANALYZING ENERGY CONSUMPTION OF ELASTIC HPC APPLICATIONS IN THE CLOUD Gustavo Rostirolla, Vinicius Facco Rodrigues, Rodrigo da Rosa Righi, and Cristiano Andr da Costa Contact: grostirolla1@gmail.com SUMMARY Introduction Energy
SLIDE 2
SLIDE 3
INTRODUCTION
- Elasticity can be a double-edged sword involving
performance and energy consumption;
Time Resources Energy
A user can achieve a good performance considering the time to execute its application, but using a large amount of resources, resulting in a waste of energy.
SLIDE 4
INTRODUCTION
- Measuring performance and energy consumption
accurately are not easy tasks;
Administrators can suffer with resource sharing among the users, besides a waste on energy consumption. Can be measured by the server. But how much power each user is consuming in a moment?
SLIDE 5
MODEL
- Deploying energy sensors or wattmeters can be
costly if not done at the time the whole infrastructure is set up ( besides being time consuming as the infrastructure scales up);
- We present an elastic energy consumption model
which gives data about energy when executing HPC applications in elastic-based cloud environments.
SLIDE 6
MODEL
The proposed model extracts energy consumption data from a malleable infrastructure of resources, enabling relationships among energy consumption, resource consumption and performance.
- 1. Collect samples of resource usage, and the machine energy
consumption using a smart power meter;
- 2. Perform regression methods to generate the energy model;
- 3. Test the model in a different set of data collected from another
homogeneous machine.
SLIDE 7
MODEL
- We obtained a mean and median accuracy of
97.15% and 97.72%, respectively.
10 20 30 40 50 60 70 80 1000 2000 3000 4000 5000 6000 7000 8000 Instantaneous Power Consumption (W) Sample Measured Power Predicted Power
SLIDE 8
MODEL
- Considers the cloud elasticity;
- Can measure shared resources power
consumption;
- Scales as the infrastructure grows up
(homogeneous).
2 VMs 4 VMs 2 VMs 4 VMs 8 VMs
SLIDE 9
MODEL
f(CPU, Memory) = α + β × CPU + δ × Memory
α represents an IDLE power consumption. β and δ represent the variable power consumption determined by the amount of resources that is used in the moment Energy consumption of machine m according to the logged CPU and memory in a instant i
MC(m, i) = f(CPU(m, i), Memory(m, i))
SLIDE 10
MODEL
ETC(t) =
Machines
X
i=0
MC(i, t) × x ⇢ x = 0 if machine i is not active in the instant t; x = 1 if machine i is active in the instant t.
TC(t) =
t
X
i=0
ETC(i) 0 ≤ t ≤ TotalApplicationTime
ETC calculates the total power consumption of all machines allocated in an instant t, taking into account elasticity using the previous equation MC TC calculates the total energy consumption using the previous equation ETC in a determine time interval.
SLIDE 11
MODEL
NEC(z) =
AppT ime
X
i=0
ETC(i) ⇥ y ⇢ y = 0 if in instant i the total of active machines 6= z; y = 1 if in instant i the total of active machines = z.
Presents the application power consumption when employing an specific amount of nodes represented by z. It allows the analysis of how much energy has been spent using a specific amount of nodes during the application execution
2 VMs 4 VMs 6 VMs 8 VMs 10 VMs
100 200 300 400 500 30 50 Energy (kJ)
SLIDE 12
METHODOLOGY
VM
Master
AutoElastic Manager AutoElastic Cloud S M S Node 0 M S Master process Slave process VM0 VMc-1 S S Node 1 VMc VM2c-1 S S Node 2 VM2c VM3c-1 S S Node m-1 VM (m-1)c VMn-1 Area for Data Share Cloud Front-End Application Virtual Machines Computational Resources Interconnection Network SSH Connection and Cloud-supported Application Program Interface (API) (b)
- 6 Nodes - 1 FrontEnd 5 Computing
- 2.9 GHz Dual Core
- 4 GB RAM
- 100 Mbps
SLIDE 13
METHODOLOGY
1 2 3 4 5 6 7 8 9 10 1 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 Number of subintervals [load(x)] x 100000 Iteration Constant Ascending Descending Wave
10 20 30 40 50 60 70 80 300 600 900 1200 1500 1800 2100 2400 2700 3000 3300 3600 3900 4200 Power Consumption (W) Time (seconds) Constant Ascending Descending Wave Load Pattern
Power Consumption of a Single Node Varying the Load Pattern
SLIDE 14
RESULTS ANALYSIS
100 200 300 400 500 30 70 50 70 30 90 50 90 Energy (kJ) Thresholds (lower / upper)
Constant
100 200 300 400 500 30 70 50 70 30 90 50 90 Energy (kJ) Thresholds (lower / upper)
Ascending Descending Wave
100 200 300 400 500 30 70 50 70 30 90 50 90 Energy (kJ) Thresholds (lower / upper) 100 200 300 400 500 30 70 50 70 30 90 50 90 Energy (kJ) Thresholds (lower / upper)
2 VMs 4 VMs 6 VMs 8 VMs 10 VMs
SLIDE 15
RESULTS ANALYSIS
- (i) Host allocation;
(ii) Virtual machines booting; (iii) Processing stop to incorporate new resources; (iv)Host Deallocation.
SLIDE 16
CONCLUSION
- A model estimates energy consumption based on CPU and
Memory traces with mean and median accuracy of 97.15% and 97.72%;
- Equations to analyze HPC application power consumption on
elastic cloud environment;
- Best energy saving with a threshold close to 90%;
- Worst energy saving with an upper threshold equal to 70%,
but it reaches the best performance rates.
SLIDE 17
- Extend the proposed model to