|
How much power is needed for a billion-thread high-throughput server?
Zhiwei XU
Front Comput Sci. 2012, 6 (4): 339-346.
https://doi.org/10.1007/s11704-012-2071-5
With the advent of Internet services, big data and cloud computing, high-throughput computing has generated much research interest, especially on high-throughput cloud servers. However, three basic questions are still not satisfactorily answered: (1) What are the basic metrics (what throughput and high-throughput of what)? (2) What are the main factors most beneficial to increasing throughput? (3) Are there any fundamental constraints and how high can the throughput go? This article addresses these issues by utilizing the fifty-year progress in Little’s law, to reveal three fundamental relations among the seven basic quantities of throughput (λ), number of active threads (L), waiting time (W), system power (P), thread energy (E), Watts per thread ω, threads per Joule θ. In addition to Little’s law L = λW, we obtain P = λE and λ = Lωθ, under reasonable assumptions. These equations help give a first order estimation of performance and power consumption targets for billion-thread cloud servers.
References |
Related Articles |
Metrics
|
|
CloudRank-D: benchmarking and ranking cloud computing systems for data processing applications
Chunjie LUO, Jianfeng ZHAN, Zhen JIA, Lei WANG, Gang LU, Lixin ZHANG, Cheng-Zhong XU, Ninghui SUN
Front Comput Sci. 2012, 6 (4): 347-362.
https://doi.org/10.1007/s11704-012-2118-7
With the explosive growth of information, more and more organizations are deploying private cloud systems or renting public cloud systems to process big data. However, there is no existing benchmark suite for evaluating cloud performance on the whole system level. To the best of our knowledge, this paper proposes the first benchmark suite CloudRank-D to benchmark and rank cloud computing systems that are shared for running big data applications.We analyze the limitations of previous metrics, e.g., floating point operations, for evaluating a cloud computing system, and propose two simple metrics: data processed per second and data processed per Joule as two complementary metrics for evaluating cloud computing systems. We detail the design of CloudRank-D that considers representative applications, diversity of data characteristics, and dynamic behaviors of both applications and system software platforms. Through experiments, we demonstrate the advantages of our proposed metrics. In several case studies, we evaluate two small-scale deployments of cloud computing systems using CloudRank-D.
References |
Related Articles |
Metrics
|
|
Dynamic cache partitioning based on hot page migration
Xiaolin WANG, Xiang WEN, Yechen LI, Zhenlin WANG, Yingwei LUO, Xiaoming LI
Front Comput Sci. 2012, 6 (4): 363-372.
https://doi.org/10.1007/s11704-012-2099-6
Static cache partitioning can reduce interapplication cache interference and improve the composite performance of a cache-polluted application and a cachesensitive application when they run on cores that share the last level cache in the same multi-core processor. In a virtualized system, since different applications might run on different virtual machines (VMs) in different time, it is inapplicable to partition the cache statically in advance. This paper proposes a dynamic cache partitioning scheme that makes use of hot page detection and page migration to improve the composite performance of co-hosted virtual machines dynamically according to prior knowledge of cache-sensitive applications. Experimental results show that the overhead of our page migration scheme is low, while in most cases, the composite performance is an improvement over free composition.
References |
Related Articles |
Metrics
|
|
Energy-aware application performance management in virtualized data centers
Hui CHEN, Ping LU, Pengcheng XIONG, Cheng-Zhong XU, Zhiping WANG
Front Comput Sci. 2012, 6 (4): 373-387.
https://doi.org/10.1007/s11704-012-2107-x
Both performance and energy cost are important concerns for current data center operators. Traditionally, however, IT and mechanical engineers have separately optimized the cyber and physical aspects of data center operations. This paper considers both of these aspects with the eventual goal of developing performance and power management techniques that operate holistically to control the entire cyber-physical complex of data center installations. Toward this end, we propose a balance of payments model for holistic power and performance management. As an example of coordinated cyber-physical system management, the energyaware cyber-physical system (EaCPS) uses an application controller on the cyber side to guarantee application performance, and on the physical side, it utilizes electric currentaware capacity management (CACM) to smartly place executables to reduce the energy consumption of each chassis present in a data center rack. A web application, representative of a multi-tier web site, is used to evaluate the performance of the controller on the cyber side, the CACM control on the physical side, and the holistic EaCPS methods in a mid-size instrumented data center. Results indicate that coordinated EaCPS outperforms separate cyber and physical control modules.
References |
Related Articles |
Metrics
|
|
Towards architecture-based management of platforms in the cloud
Gang HUANG, Xing CHEN, Ying ZHANG, Xiaodong ZHANG
Front Comput Sci. 2012, 6 (4): 388-397.
https://doi.org/10.1007/s11704-012-2100-4
System management is becoming increasingly complex and brings high costs, especially with the advent of cloud computing. Cloud computing involves numerous platforms often of virtual machines (VMs) and middleware has to be managed to make the whole system work costeffectively after an application is deployed. In order to reduce management costs, in particular for the manual activities, many computer programs have been developed remove or reduce the complexity and difficulty of system mamnagement. These programs are usually hard-coded in languages like Java and C++, which bring enough capability and flexibility but also cause high programming effort and cost. This paper proposes an architecture for developing management programs in a simple but powerful way. First of all, the manageability of a given platform (via APIs, configuration files, and scripts) is abstracted as a runtime model of the platform’s software architecture, which can automatically and immediately propagate any observable runtime changes of the target platforms to the corresponding architecture models, and vice versa. The management programs are developed using modeling languages, instead of those relatively low-level programming languages. Architecture-level management programs bring many advantages related to performance, interoperability, reusability, and simplicity. An experiment on a real-world cloud deployment and comparisonwith traditional programming language approaches demonstrate the feasibility, effectiveness, and benefits of the new model based approach for management program development.
References |
Related Articles |
Metrics
|
|
An efficient method for privacy preserving location queries
Yubao LIU, Xiuwei CHEN, Zhan LI, Zhijie LI, Raymond Chi-Wing WONG
Front Comput Sci. 2012, 6 (4): 409-420.
https://doi.org/10.1007/s11704-012-2838-8
Recently, the issue of privacy preserving location queries has attracted much research. However, there are few works focusing on the tradeoff between location privacy preservation and location query information collection. To tackle this kind of tradeoff, we propose the privacy persevering location query (PLQ), an efficient privacy preserving location query processing framework. This framework can enable the location-based query without revealing user location information. The framework can also facilitate location-based service providers to collect some information about the location based query, which is useful in practice. PLQ consists of three key components, namely, the location anonymizer at the client side, the privacy query processor at the server side, and an additional trusted third party connecting the client and server. The location anonymizer blurs the user location into a cloaked area based on a map-hierarchy. The map-hierarchy contains accurate regions that are partitioned according to real landforms. The privacy query processor deals with the requested nearest-neighbor (NN) location based query. A new convex hull of polygon (CHP) algorithm is proposed for nearest-neighbor queries using a polygon cloaked area. The experimental results show that our algorithms can efficiently process location based queries.
References |
Related Articles |
Metrics
|
|
A qualitative and quantitative study of color emotion using valence-arousal
Shangfei WANG, Rui DING
Front Comput Sci. 2012, 6 (4): 469-476.
https://doi.org/10.1007/s11704-012-0154-y
This paper describes qualitative and quantitative analysis of color emotion dimension expression using a standard device-independent colorimetric system. To collect color emotion data, 20 subjects are required to report their emotion responses, using a valence-arousal emotion model, to 52 color samples that are chosen from CIELAB Lch color spaces. Qualitative analysis, including analysis of variance (ANOVA), Pearson’s correlation and Spearman’s rank correlation, shows that significant differences exist between responses to achromatic and chromatic stimuli, but there are no significant differences between chromatic samples. There is a positive linear relationship between lightness/chroma and valence-arousal dimensions. Subsequently, several classic predictors are used to quantitatively predict emotion induced by color attributes. Furthermore, several explicit color emotion models are developed by using multiple linear regression with stepwise and pace regression. Experimental results show that chroma and lightness have stronger effects on emotions than hue, which is consistent with our qualitative results and other psychological studies. Arousal has greater predictive value than valence.
References |
Related Articles |
Metrics
|
|
A uniform solution to the independent set problem through tissue P systems with cell separation
Xingyi ZHANG, Xiangxiang ZENG, Bin LUO, Zheng ZHANG
Front Comput Sci. 2012, 6 (4): 477-488.
https://doi.org/10.1007/s11704-012-1054-x
Membrane computing is an emergent branch of natural computing, which is inspired by the structure and the functioning of living cells, as well as the organization of cells in tissues, organs, and other higher order structures. Tissue P systems are a class of the most investigated computing models in the framework of membrane computing, especially in the aspect of efficiency. To generate an exponential resource in a polynomial time, cell separation is incorporated into such systems, thus obtaining so called tissue P systems with cell separation. In this work, we exploit the computational efficiency of this model and construct a uniform family of such tissue P systems for solving the independent set problem, a well-known NP-complete problem, by which an efficient solution can be obtained in polynomial time.
References |
Related Articles |
Metrics
|
12 articles
|