Informatik
Refine
Document Type
- Journal article (6) (remove)
Language
- English (6)
Is part of the Bibliography
- yes (6)
Institute
- Informatik (6)
Publisher
The cloud evolved into an attractive execution environment for parallel applications, which make use of compute resources to speed up the computation of large problems in science and industry. Whereas Infrastructure as a Service (IaaS) offerings have been commonly employed, more recently, serverless computing emerged as a novel cloud computing paradigm with the goal of freeing developers from resource management issues. However, as of today, serverless computing platforms are mainly used to process computations triggered by events or user requests that can be executed independently of each other and benefit from on-demand and elastic compute resources as well as per-function billing. In this work, we discuss how to employ serverless computing platforms to operate parallel applications. We specifically focus on the class of parallel task farming applications and introduce a novel approach to free developers from both parallelism and resource management issues. Our approach includes a proactive elasticity controller that adapts the physical parallelism per application run according to user-defined goals. Specifically, we show how to consider a user-defined execution time limit after which the result of the computation needs to be present while minimizing the associated monetary costs. To evaluate our concepts, we present a prototypical elastic parallel system architecture for self-tuning serverless task farming and implement two applications based on our framework. Moreover, we report on performance measurements for both applications as well as the prediction accuracy of the proposed proactive elasticity control mechanism and discuss our key findings.
Elasticity is considered to be the most beneficial characteristic of cloud environments, which distinguishes the cloud from clusters and grids. Whereas elasticity has become mainstream for web-based, interactive applications, it is still a major research challenge how to leverage elasticity for applications from the high-performance computing (HPC) domain, which heavily rely on efficient parallel processing techniques. In this work, we specifically address the challenges of elasticity for parallel tree search applications. Well-known meta-algorithms based on this parallel processing technique include branch-and-bound and backtracking search. We show that their characteristics render static resource provisioning inappropriate and the capability of elastic scaling desirable. Moreover, we discuss how to construct an elasticity controller that reasons about the scaling behavior of a parallel system at runtime and dynamically adapts the number of processing units according to user-defined cost and efficiency thresholds. We evaluate a prototypical elasticity controller based on our findings by employing several benchmarks for parallel tree search and discuss the applicability of the proposed approach. Our experimental results show that, by means of elastic scaling, the performance can be controlled according to user-defined thresholds, which cannot be achieved with static resource provisioning.
Parallel applications are the computational backbone of major industry trends and grand challenges in science. Whereas these applications are typically constructed for dedicated High Performance Computing clusters and supercomputers, the cloud emerges as attractive execution environment, which provides on-demand resource provisioning and a pay-per-use model. However, cloud environments require specific application properties that may restrict parallel application design. As a result, design trade-offs are required to simultaneously maximize parallel performance and benefit from cloud-specific characteristics.
In this paper, we present a novel approach to assess the cloud readiness of parallel applications based on the design decisions made. By discovering and understanding the implications of these parallel design decisions on an application’s cloud readiness, our approach supports the migration of parallel applications to the cloud.We introduce an assessment procedure, its underlying meta model, and a corresponding instantiation to structure this multi-dimensional design space. For evaluation purposes, we present an extensive case study comprising three parallel applications and discuss their cloud readiness based on our approach.
In this paper, we deal with optimizing the monetary costs of executing parallel applications in cloud-based environments. Specifically, we investigate on how scalability characteristics of parallel applications impact the total costs of computations. We focus on a specific class of irregularly structured problems, where the scalability typically depends on the input data. Consequently, dynamic optimization methods are required for minimizing the costs of computation. For quantifying the total monetary costs of individual parallel computations, the paper presents a cost model that considers the costs for the parallel infrastructure employed as well as the costs caused by delayed results. We discuss a method for dynamically finding the number of processors for which the total costs based on our cost model are minimal. Our extensive experimental evaluation gives detailed insights into the performance characteristics of our approach.
With on-demand access to compute resources, pay-per-use, and elasticity, the cloud evolved into an attractive execution environment for High Performance Computing (HPC). Whereas elasticity, which is often referred to as the most beneficial cloud-specific property, has been heavily used in the context of interactive (multi-tier) applications, elasticity-related research in the HPC domain is still in its infancy. Existing parallel computing theory as well as traditional metrics to analytically evaluate parallel systems do not comprehensively consider elasticity, i.e., the ability to control the number of processing units at runtime. To address these issues, we introduce a conceptual framework to understand elasticity in the context of parallel systems, define the term elastic parallel system, and discuss novel metrics for both elasticity control at runtime as well as the ex post performance evaluation of elastic parallel systems. Based on the conceptual framework, we provide an in depth analysis of existing research in the field to describe the state-of-the art and compile our findings into a research agenda for future research on elastic parallel systems.
Container virtualization evolved into a key technology for deployment automation in line with the DevOps paradigm. Whereas container management systems facilitate the deployment of cloud applications by employing container based artifacts, parts of the deployment logic have been applied before to build these artifacts. Current approaches do not integrate these two deployment phases in a comprehensive manner. Limited knowledge on application software and middleware encapsulated in container-based artifacts leads to maintainability and configuration issues. Besides, the deployment of cloud applications is based on custom orchestration solutions leading to lock in problems. In this paper, we propose a two-phase deployment method based on the TOSCA standard. We present integration concepts for TOSCA-based orchestration and deployment automation using container-based artifacts. Our two-phase deployment method enables capturing and aligning all the deployment logic related to a software release leading to better maintainability. Furthermore, we build a container management system, which is composed of a TOSCA-based orchestrator on Apache Mesos, to deploy container-based cloud applications automatically.