Thursday, 14 December 2017

Session 10: Posters

LIMBO: A Tool for Modeling Variable Load Intensities

Authors:

Jóakim v. Kistowski (Karlsruhe Institute of Technology)
Nikolas Herbst (Karlsruhe Institute of Technology)
Samuel Kounev (Karlsruhe Institute of Technology)

Abstract:

Modern software systems are expected to deliver reliable performance under highly variable load intensities while at the same time making efficient use of dynamically allocated resources. Conventional benchmarking frameworks provide limited support for emulating such highly variable and dynamic load profiles and workload scenarios. Industrial benchmarks typically use workloads with constant or stepwise increasing load intensity, or they simply replay recorded workload traces. In this paper, we present LIMBO - an Eclipse- based tool for modeling variable load intensity profiles based on the Descartes Load Intensity Model [3] as an underlying modeling formalism.

DOI: 10.1145/2568088.2576092

Full text: PDF

[#][]

LibReDE: A Library for Resource Demand Estimation

Authors:

Simon Spinner (Karlsruhe Institute of Technology (KIT))
Giuliano Casale (Imperial College London)
Xiaoyun Zhu (VMware, Inc.)
Samuel Kounev (Karlsruhe Institute of Technology (KIT))

Abstract:

When creating a performance model, it is necessary to quantify the amount of resources consumed by an application serving individual requests. In distributed enterprise systems, these resource demands usually cannot be observed directly, their estimation is a major challenge. Different statistical approaches to resource demand estimation based on monitoring data have been proposed, e.g., using linear regression or Kalman filtering techniques. In this paper, we present LibReDE, a library of ready-to-use implementations of approaches to resource demand estimation that can be used for online and offline analysis. It is the first publicly available tool for this task and aims at supporting performance engineers during performance model construction. The library enables the quick comparison of the estimation accuracy of different approaches in a given context and thus helps to select an optimal one.

DOI: 10.1145/2568088.2576093

Full text: PDF

[#][]

Server Efficiency Rating Tool (SERT) 1.0.2: An Overview

Authors:

Hansfried Block (Fujitsu Technology Solutions GmbH)
Jeremy A. Arnold (IBM Corporation)
John Beckett (Dell Inc.)
Sanjay Sharma (Intel Corporation)
Mike G. Tricker (Microsoft Corporation)
Kyle M. Rogers (Hewlett-Packard Company)

Abstract:

The Server Efficiency Rating Tool (SERT) has released the Standard Performance Evaluation Corporation (SPEC) and the EPA released Version 2.0 of the ENERGY STAR for Computer Servers program in early 2013 to include the mandatory use of the SERT. Other governments world-wide that are concerned with the growing power consumption of servers and datacenters are also considering adoption of the SERT. This poster-paper provides an overview of the current release of 1.0.2 version of SERT.

DOI: 10.1145/2568088.2576094

Full text: PDF

[#][]

SPECjbb2013 1.0: An Overview

Authors:

Charles Pogue (Hewlett-Packard Company)
Anil Kumar (Intel Corporation)
Douglas Tollefson (IBM Corporation)
Steve Realmuto (Oracle Corporation)

Abstract:

SPECjbb2013 [1] is an entirely new version of the industry standard benchmark for evaluating Java server business performance from Standard Performance Evaluation Corporation (SPEC) [2]. It is designed with three categories which allow multiple configurations (Composite/single host, MultiJVMs/ single host, Distributed/single or multi hosts), enabling the user to systematically analyze their system. Additionally, the status of published results is summarized and a series of research project configurations are suggested.

DOI: 10.1145/2568088.2576095

Full text: PDF

[#][]