270 likes | 412 Views
Xinying Zheng, Yu cai Michigan Technological University. Optimal Server Provisioning and Frequency Adjustment in Server Clusters. Presented by: Xinying Zheng. Outline. Introduction Related Work Optimization problem formulation Single class Multiple classes
E N D
Xinying Zheng, Yu cai Michigan Technological University Optimal Server Provisioning and Frequency Adjustment in Server Clusters Presented by: Xinying Zheng
Outline • Introduction • Related Work • Optimization problem formulation • Single class • Multiple classes • Overhead Analysis: DCP model • Performance Evaluation • Conclusion and Future Work
Motivation • The increased data centers and cluster systems consume significant amount of energy.
Motivation • The power consumption of enterprise data centers in the U.S. doubled between 2000 and 2005. And will likely triple in the next few years. • Servers consume 0.5 percent of the worlds total electricity usage, this number will increase to 2 percent by 2020.
DVS( Dynamic voltage scaling) Processor Feedback Control Memory Single Server DTM( Dynamic thermal management) Discs Non-data Movement Data Movement Performance level Storage and Database Servers Feedback Control DVC DV/FS Server Cluster DTM Long-live connected service Web and application Servers VOVF Request-response service Virtualization Wireless sensor networks Memory Network Techniques Computer networks Economic method
Syetem Assumption • All servers in the cluster are identical nodes. • Each server has two modes: active and inactive. • Operate at a number of discrete frequencies. • All the incoming requests are CPU bounded.
Performance metric modeling • Incoming request follows a heavy-tailed bounded Pareto distribution. • If we define a function: • Average job size: (1) (2) (3) (4) (5)
Request time in single server • Server processing capacity: c • Packets inter-arrival time follows exponential distribution with a mean of 1/λ. • According to Pollaczek-Khinchin formula, the average waiting time is : • Request time: (6) (7) (8) (9)
Extend to server cluster • Extend to the server-cluster mode. Using Round-Robin dispatching policy, the arrival process at each server in the cluster has rate . • Processing capacity is proportional to frequency. • Request time: (10) (11)
Power consumption modeling • Power-to-frequency relationship. • Linear model. • Cubic model: • System power consumption: (12) (13)
Optimization problem formulation • Minimizing total power consumption. • Request time threshold. • Mechanism: • VOVF: vary-on, vary-off • DFS: dynamic frequency scaling.
Optimization problem formulation (single class) • Single class: • Computation complexity is O(NM). • Complexity can be reduced to O(NM). applying a coordinated voltage scaling. (14)
Optimization problem formulation (Multiple classes) • Assuming incoming requests are classified into N classes. • The ratio of average request time between class i and j is fixed to the ratio of the corresponding differentiation parameters: • We assume class 1 is the “highest class” and set: (15)
Optimization problem formulation (Multiple classes) • Multiple classes: • Different class receive different performance. (16)
Overhead Analysis • Server transfers from inacitve to active mode. • Transition time influence the performance. • Double Control Periods(DCP) model. Double control periods
Evaluation (single class) Request time comparison between OP model and DCP model Power consumption comparison between OP model and DCP model
Evaluation (multiple classes) Request time comparison between OP model and DCP model Power consumption comparison between OP model and DCP model
Evaluation(real workload single class) Request time comparison between OP model and DCP model Power consumption comparison between OP model and DCP model
Evaluation(real workload multiple classes) Request time comparison between OP model and DCP model Power consumption comparison between OP model and DCP model
Contributions • Optimization model for power reduction in server clusters. • Single class and multiple classes. • Double control periods model to compensate the transition overhead. • Evaluate our models in real workload data trace.
Future work • Effect of dispatching strategy. • Transition overhead of frequency adjustment. • heterogeneity in data centers. • Apply our model to the real Internet web servers in the future.
Questions Thanks for your attention