Thursday, 14 December 2017

Session 6: Benchmarking 2

Session Chair: Alberto Avritzer (Siemens Corporate Research)

Resource Demand Modeling for Multi-Tier Services

Authors:

Jerry Rolia (Hewlett Packard Laboratories)
Amir Kalbasi (University of Calgary)
Diwakar Krishnamurthy (University of Calgary)
Stephen Dawson (SAP Research)

Abstract:

We present a new technique for predicting the resource demand requirements of services implemented by multi-tier systems. Accurate demand estimates are essential to ensure the efficient provisioning of services in an increasingly service-oriented world. The demand estimation technique proposed in this paper has several advantages compared with regression-based demand estimation techniques, which many practitioners employ today. In contrast to regression, it does not suffer from the problem of multicollinearity, it provides more reliable aggregate resource demand and confidence interval predictions, and it offers a measurement-based validation test. The technique can be used to support system sizing and capacity planning exercises, costing and pricing exercises, and to predict the impact of changes to a service upon different service customers.

DOI: 10.1145/1712605.1712638

Full text: PDF

[#][]

Addressing the Stranded Power Problem in Datacenters using Storage Workload Characterization

Authors:

Sriram Sankar (Microsoft Corporation)
Kushagra Vaid (Microsoft Corporation)

Abstract:

Datacenter operators face unique challenges to optimally provision power among deployed servers. Allocated server power is frequently over-provisioned and this results in stranding of available datacenter power capacity. Standardized power efficiency benchmarks like SPECpower_ssj2008 can be used for determining power allocation, in conjunction with methodologies to estimate the contribution from the disk subsystem. In this paper, we explore a trace-driven methodology for determining power contribution of the storage components. We show the benefits of this methodology as opposed to typical power provisioning used in the industry.

DOI: 10.1145/1712605.1712639

Full text: PDF

[#][]

Reducing Performance Non-determinism via Cache-aware Page Allocation Strategies

Authors:

Michal Hocko (Charles University)
Tomas Kalibera (Charles University & Purdue University)

Abstract:

Performance non-determinism in computer systems complicates evaluation, use, and even development of these systems. In performance evaluation via benchmarking and simulation, non-determinism requires long executions and more complex experiment design. Real-time systems are hard to dimension and tune with non-determinism. The slower benchmarking also slows down system development, as it takes developers longer to see performance implications of their modifications.

Cache-unaware physical page allocation in an operating system is believed to be a significant cause of non-determinism, but there is no published empirical study that would confirm it.

We provide such a study for the Linux operating system, comparing the default cache-unaware page allocation strategy to known cache-aware strategies, page coloring and bin hopping. We have implemented a framework for page allocation strategies in the Linux kernel, employed it for these two strategies, and measured the non-determinism on a large and diverse set of benchmarks. We propose a statistical technique which allows to classify different kinds of performance non-determinism and evaluate their magnitudes. Application of our technique reveals that the two strategies do reduce performance non-determinism without significantly increasing mean response time.

DOI: 10.1145/1712605.1712640

Full text: PDF

[#][]

Quantifying Load Imbalance on Virtualized Enterprise Servers

Authors:

Emmanuel Arzuaga (Northeastern University)
David R. Kaeli (Northeastern University)

Abstract:

Virtualization has been shown to be an attractive path to increase overall system resource utilization. The use of live virtual machine (VM) migration has enabled more effective sharing of system resources across multiple physical servers, resulting in an increase in overall performance. Live VM migration can be used to load balance virtualized clusters. To drive live migration, we need to be able to measure the current load imbalance. Further, we also need to accurately predict the resulting load imbalance produced by any migration.

In this paper we present a new metric that captures the load of the physical servers and is a function of the resident VMs. This metric will be used to measure load imbalance and construct a load-balancing VM migration framework. The algorithm for balancing the load of virtualized enterprise servers follows a greedy approach, inductively predicting which VM migration will yield the greatest improvement of the imbalance metric in a particular step. We compare our algorithm to the leading commercially available load balancing solution - VMware's Distributed Resource Scheduler (DRS). Our results show that when we are able to accurately measure system imbalance, we can also predict future system state. We find that we can outperform DRS and improve performance up to 5%. Our results show that our approach does not impose additional performance impact and is comparable to the virtual machine monitor overhead.

DOI: 10.1145/1712605.1712641

Full text: PDF

[#][]