- The document presents a self-adaptive capacity management scheme for internet data centers (IDCs) that aims to maximize a service provider's revenue while satisfying customers' service level agreements (SLAs).
- It uses a control interval framework to dynamically adjust VM capacity based on performance models and a cost model. Performance is estimated using queueing theory, and costs account for per-use pricing and penalties/rewards for meeting/missing SLAs.
- Experimental analysis shows the self-adaptive approach increases provider payoff and maintains application stability compared to static configurations, while different performance approximations provide similar accuracy.
Self-Adaptive SLA-Driven Capacity Management for Internet Services
1. IEEE NOMS 2006
Self-Adaptive SLA-Driven
6 April, 2006
Capacity Management for
Internet Services
Bruno Abrahao, Virgilio Almeida, Jussara Almeida
Federal University of Minas Gerais, Brazil
Alex Zhang, Dirk Beyer, Fereydoon Safai
Hewllet-Packard Labs Palo Alto, CA
2. Motivation
• IT outsourcing for Internet Services
− Contracts with a provider
− Multiple service shared Internet Data Centers (IDC)
• Providers’ challenging task
− cost effectiveness while satisfying the customers’ SLA
requirements
• Complexity
− Keep track of different application requirements, systems
characteristics, and simultaneous workload variations, as well as
(and more importantly!) to consider the business goal of the
provider
2
3. Challenges
•New customer demands
Multiple metric Probabilistic Per use service
requirements performance accounting
requirements
•Application characteristics
High workload Unexpected Application
fluctuations workload peaks Heterogeneity
• manual management •even more complex business
becomes impractical and systems models
3
4. Goal
• To present a self-adaptive capacity management
scheme for IDCs which aims at maximizing the
service revenue of the provider
− Take into account the new challenges of the modern IT
business and infra-structure
− Allows providers to offer customers flexible service
plans
− Minimize management costs for service providers
4
5. IDC Environment
•Virtualization • VMs provide admission control
mechanisms
•Transparent and flexible
capacity expansion/ contraction.
5
7. Capacity Manager Scheme
• Provides IDC configurations that maximize the business
objective of the provider
7
8. Cost Model
• Allows per-use service accounting
− Customers pay for extra capacity (than that normally required) only
when needed
• Service accounting
− performance achieved by virtual machines instead of simply
accounting for resource utilization
8
9. Cost Model
• Allows probabilistic response time requirements
P( Ri RiSLA ) i
• Allows multiple metric service level
− Throughput, subjected to a guarantee in the response time of
the processed transactions
{X | P( R R SLA ) }
9
10. Cost Model
Extra processing limit Two-level SLA contracts
- Normal operation mode
-Surge operation mode
Normal processing requirement Penalty/Reward model
Provider’s business objetive
Maximize the net result from the
penalties and rewards
10
11. Performance Model
•Based on queuing-theory Capacity allocation
decision
•application system
characteristics
Performance
•performance
requirements Model
•current workload intensity
•Throughput
•Utilization
•Response time probability
distribution
11
12. Performance Model
• Utilization and Throughput can be estimated using well-known
queuing-based formulas
• Approximations are often needed to estimate Response time
probability distribution
E[ Ri ]
− Markov P( Ri R i
SLA
) SLA
Ri
var[Ri ]
− Chebyshev P( Ri R SLA
) SLA
( R E[ Ri ])2
RiS LA( f i / E[ Si ])(1 i )
− Percentile (M/M/1) P( Ri R i
SLA
)e
12
13. Optimization model
Provider’s business objective
Cost
Model
{
Perf.
Model
Capacity
{
allocation
13
14. Experimental Analysis
• Self-adaptive versus static configuration
− Examine the resulting provider’s payoff
− Examine whether performance requirements are met and queue
stability is maintained
• Compare the degree of accuracy provided by each of the
performance approximations
• how
− Simulate and analyze the behavior of two competing applications
that receive different workloads levels over time
14
19. Conclusions
• The self-adaptive capacity management model
with any of the approximations is able to
− increase the business potential of the provider
− Higher payoffs
− maintain the application stability
− Stable service queues
− Response time requirement satisfaction
− Markov’s approximation overestimates capacity needs
− Chebyshev e Percentile result in a equivalent degree of
precision in M/M/1 model
• Allows for the new challenges of the problem
19
20. Time for questions
IEEE NOMS 2006
Self-Adaptive SLA-Driven 6 April, 2006
Capacity Management for
Internet Services
Bruno Abrahao, Virgilio Almeida, Jussara Almeida
Universidade Federal de Minas Gerais, Brazil
Alex Zhang, Dirk Beyer, Fereydoon Safai
Hewllet-Packard Labs Palo Alto, CA