arXiv subjects

Jianfeng Zhan

Publications and source records attributed to Jianfeng Zhan.

104 records · Page 6Linked to original sources

arXiv2011

Automatic Performance Debugging of SPMD-style Parallel Programs

The simple program and multiple data (SPMD) programming model is widely used for both high performance computing and Cloud computing. In this paper, we design and implement an innovative system, AutoAnalyzer, that automates the process of debugging performance problems of SPMD-style parallel programs, including data collection, performance behavior analysis, locating bottlenecks, and uncovering their root causes. AutoAnalyzer is unique in terms of two features: first, without any apriori knowledge, it automatically locates bottlenecks and uncovers their root causes for performance optimization; second, it is lightweight in terms of the size of performance data to be collected and analyzed. Our contributions are three-fold: first, we propose two effective clustering algorithms to investigate the existence of performance bottlenecks that cause process behavior dissimilarity or code region behavior disparity, respectively; meanwhile, we present two searching algorithms to locate bottlenecks; second, on a basis of the rough set theory, we propose an innovative approach to automatically uncovering root causes of bottlenecks; third, on the cluster systems with two different configurations, we use two production applications, written in Fortran 77, and one open source code-MPIBZIP2 (http://compression.ca/mpibzip2/), written in C++, to verify the effectiveness and correctness of our methods. For three applications, we also propose an experimental approach to investigating the effects of different metrics on locating bottlenecks.

cs.DC↗

arXiv2010

PowerTracer: Tracing requests in multi-tier services to save cluster power consumption

As energy proportional computing gradually extends the success of DVFS (Dynamic voltage and frequency scaling) to the entire system, DVFS control algorithms will play a key role in reducing server clusters' power consumption. The focus of this paper is to provide accurate cluster-level DVFS control for power saving in a server cluster. To achieve this goal, we propose a request tracing approach that online classifies the major causal path patterns of a multi-tier service and monitors their performance data as a guide for accurate DVFS control. The request tracing approach significantly decreases the time cost of performance profiling experiments that aim to establish the empirical performance model. Moreover, it decreases the controller complexity so that we can introduce a much simpler feedback controller, which only relies on the single-node DVFS modulation at a time as opposed to varying multiple CPU frequencies simultaneously. Based on the request tracing approach, we present a hybrid DVFS control system that combines an empirical performance model for fast modulation at different load levels and a simpler feedback controller for adaption. We implement a prototype of the proposed system, called PowerTracer, and conduct extensive experiments on a 3-tier platform. Our experimental results show that PowerTracer outperforms its peer in terms of power saving and system performance.

cs.DC↗

arXiv2010

Precise, Scalable and Online Request Tracing for Multi-tier Services of Black Boxes

As more and more multi-tier services are developed from commercial off-the-shelf components or heterogeneous middleware without source code available, both developers and administrators need a request tracing tool to (1) exactly know how a user request of interest travels through services of black boxes; (2) obtain macro-level user request behavior information of services without the necessity of inundating within massive logs. Previous research efforts either accept imprecision of probabilistic correlation methods or present precise but unscalable tracing approaches that have to collect and analyze large amount of logs; Besides, previous precise request tracing approaches of black boxes fail to propose macro-level abstractions that enables debugging performance-in-the-large, and hence users have to manually interpret massive logs. This paper introduces a precise, scalable and online request tracing tool, named PreciseTracer, for multi-tier services of black boxes. Our contributions are four-fold: first, we propose a precise request tracing algorithm for multi-tier services of black boxes, which only uses application-independent knowledge; second, we respectively present micro-level and macro-level abstractions: component activity graphs and dominated causal path patterns to represent causal paths of each individual request and repeatedly executed causal paths that account for significant fractions; third, we present two mechanisms: tracing on demand and sampling to significantly increase system scalability; fourth, we design and implement an online request tracing tool. PreciseTracer's fast response, low overhead and scalability make it a promising tracing tool for large-scale production systems.

cs.DC↗

arXiv2010

PhoenixCloud: Provisioning Resources for Heterogeneous Workloads in Cloud Computing

As more and more service providers choose Cloud platforms, which is provided by third party resource providers, resource providers needs to provision resources for heterogeneous workloads in different Cloud scenarios. Taking into account the dramatic differences of heterogeneous workloads, can we coordinately provision resources for heterogeneous workloads in Cloud computing? In this paper we focus on this important issue, which is investigated by few previous work. Our contributions are threefold: (1) we respectively propose a coordinated resource provisioning solution for heterogeneous workloads in two typical Cloud scenarios: first, a large organization operates a private Cloud for two heterogeneous workloads; second, a large organization or two service providers running heterogeneous workloads revert to a public Cloud; (2) we build an agile system PhoenixCloud that enables a resource provider to create coordinated runtime environments on demand for heterogeneous workloads when they are consolidated on a Cloud site; and (3) A comprehensive evaluation has been performed in experiments. For two typical heterogeneous workload traces: parallel batch jobs and Web services, our experiments show that: a) in a private Cloud scenario, when the throughput is almost same like that of a dedicated cluster system, our solution decreases the configuration size of a cluster by about 40%; b) in a public Cloud scenario, our solution decreases not only the total resource consumption, but also the peak resource consumption maximally to 31% with respect to that of EC2 +RightScale solution.

cs.DC↗

arXiv2010

Phoenix Cloud: Consolidating Different Computing Loads on Shared Cluster System for Large Organization

Different departments of a large organization often run dedicated cluster systems for different computing loads, like HPC (high performance computing) jobs or Web service applications. In this paper, we have designed and implemented a cloud management system software Phoenix Cloud to consolidate heterogeneous workloads from different departments affiliated to the same organization on the shared cluster system. We have also proposed cooperative resource provisioning and management policies for a large organization and its affiliated departments, running HPC jobs and Web service applications, to share the consolidated cluster system. The experiments show that in comparison with the case that each department operates its dedicated cluster system, Phoenix Cloud significantly decreases the scale of the required cluster system for a large organization, improves the benefit of the scientific computing department, and at the same time provisions enough resources to the other department running Web services with varying loads.

cs.DC↗

arXiv2010

In Cloud, Do MTC or HTC Service Providers Benefit from the Economies of Scale?

In this paper, we intend to answer one key question to the success of cloud computing: in cloud, do many task computing (MTC) or high throughput computing (HTC) service providers, which offer the corresponding computing service to end users, benefit from the economies of scale? Our research contributions are three-fold: first, we propose an innovative usage model, called dynamic service provision (DSP) model, for MTC or HTC service providers. In the DSP model, the resource provider provides the service of creating and managing runtime environments for MTC or HTC service providers, and consolidates heterogeneous MTC or HTC workloads on the cloud platform; second, according to the DSP model, we design and implement DawningCloud, which provides automatic management for heterogeneous workloads; third, a comprehensive evaluation of DawningCloud has been performed in an emulatation experiment. We found that for typical workloads, in comparison with the previous two cloud solutions, DawningCloud saves the resource consumption maximally by 46.4% (HTC) and 74.9% (MTC) for the service providers, and saves the total resource consumption maximally by 29.7% for the resource provider. At the same time, comparing with the traditional solution that provides MTC or HTC services with dedicated systems, DawningCloud is more cost-effective. To this end, we conclude that for typical MTC and HTC workloads, on the cloud platform, MTC and HTC service providers and the resource provider can benefit from the economies of scale.

cs.DC↗

arXiv2010

Scalable Group Management in Large-Scale Virtualized Clusters

To save cost, recently more and more users choose to provision virtual machine resources in cluster systems, especially in data centres. Maintaining a consistent member view is the foundation of reliable cluster managements, and it also raises several challenge issues for large scale cluster systems deployed with virtual machines (which we call virtualized clusters). In this paper, we introduce our experiences in design and implementation of scalable member view management on large-scale virtual clusters. Our research contributions are three-fold: 1) we propose a scalable and reliable management infrastructure that combines a peer-to-peer structure and a hierarchy structure to maintain a consistent member view in virtual clusters; 2) we present a light-weighted group membership algorithm that can reach the consistent member view within a single round of message exchange; and 3) we design and implement a scalable membership service that can provision virtual machines and maintain a consistent member view in virtual clusters. Our work is verified on Dawning 5000A, which ranked No.10 of Top 500 super computers in November, 2008.

cs.DC↗

arXiv2010

In Cloud, Can Scientific Communities Benefit from the Economies of Scale?

The basic idea behind Cloud computing is that resource providers offer elastic resources to end users. In this paper, we intend to answer one key question to the success of Cloud computing: in Cloud, can small or medium-scale scientific computing communities benefit from the economies of scale? Our research contributions are three-fold: first, we propose an enhanced scientific public cloud model (ESP) that encourages small- or medium-scale organizations to rent elastic resources from a public cloud provider; second, on a basis of the ESP model, we design and implement the DawningCloud system that can consolidate heterogeneous scientific workloads on a Cloud site; third, we propose an innovative emulation methodology and perform a comprehensive evaluation. We found that for two typical workloads: high throughput computing (HTC) and many task computing (MTC), DawningCloud saves the resource consumption maximally by 44.5% (HTC) and 72.6% (MTC) for service providers, and saves the total resource consumption maximally by 47.3% for a resource provider with respect to the previous two public Cloud solutions. To this end, we conclude that for typical workloads: HTC and MTC, DawningCloud can enable scientific communities to benefit from the economies of scale of public Clouds.

cs.DC↗

arXiv2010

PhoenixCloud: Provisioning Resources for Heterogeneous Cloud Workloads

As more and more service providers choose Cloud platforms, a resource provider needs to provision resources and supporting runtime environments (REs) for heterogeneous workloads in different scenarios. Previous work fails to resolve this issue in several ways: (1) it fails to pay attention to diverse RE requirements, and does not enable creating coordinated REs on demand; (2) few work investigates coordinated resource provisioning for heterogeneous workloads. In this paper, our contributions are three-fold: (1) we present an RE agreement that expresses diverse RE requirements, and build an innovative system PhoenixCloud that enables a resource provider to create REs on demand according to RE agreements; (2) we propose two coordinated resource provisioning solutions for heterogeneous workloads in two typical Cloud scenarios: first, a large organization operates a private Cloud for two heterogeneous workloads; second, a large organization or two service providers running heterogeneous workloads revert to a public Cloud; and (3) A comprehensive evaluation has been performed in experiments. For typical workload traces of parallel batch jobs and Web services, our experiments show that: a) In the first Cloud scenario, when the throughput is almost same like that of a dedicated cluster system, our solution decreases the configuration size of cluster by about 40%; b) in the second scenario, our solution decreases not only the total resource consumption, but also the peak resource consumption maximally to 31% with respect to that of EC2 + RightScale solution.

cs.DC↗

arXiv2010

Decreasing log data of multi-tier services for effective request tracing

Previous work shows request tracing systems help understand and debug the performance problems of multi-tier services. However, for large-scale data centers, more than hundreds of thousands of service instances provide online service at the same time. Previous work such as white-box or black box tracing systems will produce large amount of log data, which would be correlated into large quantities of causal paths for performance debugging. In this paper, we propose an innovative algorithm to eliminate valueless logs of multitiers services. Our experiment shows our method filters 84% valueless causal paths and is promising to be used in large-scale data centers.

cs.DC↗

arXiv2010

Precise Request Tracing and Performance Debugging for Multi-tier Services of Black Boxes

As more and more multi-tier services are developed from commercial components or heterogeneous middleware without the source code available, both developers and administrators need a precise request tracing tool to help understand and debug performance problems of large concurrent services of black boxes. Previous work fails to resolve this issue in several ways: they either accept the imprecision of probabilistic correlation methods, or rely on knowledge of protocols to isolate requests in pursuit of tracing accuracy. This paper introduces a tool named PreciseTracer to help debug performance problems of multi-tier services of black boxes. Our contributions are two-fold: first, we propose a precise request tracing algorithm for multi-tier services of black boxes, which only uses application-independent knowledge; secondly, we present a component activity graph abstraction to represent causal paths of requests and facilitate end-to-end performance debugging. The low overhead and tolerance of noise make PreciseTracer a promising tracing tool for using on production systems.

cs.DC↗

arXiv2010

Automatic Performance Debugging of SPMD Parallel Programs

Automatic performance debugging of parallel applications usually involves two steps: automatic detection of performance bottlenecks and uncovering their root causes for performance optimization. Previous work fails to resolve this challenging issue in several ways: first, several previous efforts automate analysis processes, but present the results in a confined way that only identifies performance problems with apriori knowledge; second, several tools take exploratory or confirmatory data analysis to automatically discover relevant performance data relationships. However, these efforts do not focus on locating performance bottlenecks or uncovering their root causes. In this paper, we design and implement an innovative system, AutoAnalyzer, to automatically debug the performance problems of single program multi-data (SPMD) parallel programs. Our system is unique in terms of two dimensions: first, without any apriori knowledge, we automatically locate bottlenecks and uncover their root causes for performance optimization; second, our method is lightweight in terms of size of collected and analyzed performance data. Our contribution is three-fold. First, we propose a set of simple performance metrics to represent behavior of different processes of parallel programs, and present two effective clustering and searching algorithms to locate bottlenecks. Second, we propose to use the rough set algorithm to automatically uncover the root causes of bottlenecks. Third, we design and implement the AutoAnalyzer system, and use two production applications to verify the effectiveness and correctness of our methods. According to the analysis results of AutoAnalyzer, we optimize two parallel programs with performance improvements by minimally 20% and maximally 170%.

cs.DC↗

arXiv2009

Similarity Analysis in Automatic Performance Debugging of SPMD Parallel Programs

Different from sequential programs, parallel programs possess their own characteristics which are difficult to analyze in the multi-process or multi-thread environment. This paper presents an innovative method to automatically analyze the SPMD programs. Firstly, with the help of clustering method focusing on similarity analysis, an algorithm is designed to locate performance problems in parallel programs automatically. Secondly a Rough Set method is used to uncover the performance problem and provide the insight into the micro-level causes. Lastly, we have analyzed a production parallel application to verify the effectiveness of our method and system.

cs.DC↗

arXiv2009

Multidimensional Analysis of System Logs in Large-scale Cluster Systems

It is effective to improve the reliability and availability of large-scale cluster systems through the analysis of failures. Existed failure analysis methods understand and analyze failures from one or few dimension. The analysis results are partial and with less precision because of the limitation of data source. This paper presents multidimensional analysis based on graph mining to analyze multi-source system logs, which is a promising failure analysis method to get more complete and precise failure knowledge.

cs.DC↗