Cloud Computing is one of the hottest topics researched today, with the objective of taking advantage of data center computational resources. Hardware and software virtualization make the environment scalable, redundant, and lower cost. This paper intends to characterize scientific and transactional applications in Cloud infrastructures - IaaS, identifying the best virtual machine configuration in terms of the optimal processor allocation for executing parallel and distributed applications. Through the study, it was achieved ~3 times improvements of the elapse time on scientific and transactional applications.
{"title":"Characterization of Scientific and Transactional Applications under Multi-core Architectures on Cloud Computing Environment","authors":"Denis R. Ogura, E. T. Midorikawa","doi":"10.1109/CSE.2010.47","DOIUrl":"https://doi.org/10.1109/CSE.2010.47","url":null,"abstract":"Cloud Computing is one of the hottest topics researched today, with the objective of taking advantage of data center computational resources. Hardware and software virtualization make the environment scalable, redundant, and lower cost. This paper intends to characterize scientific and transactional applications in Cloud infrastructures - IaaS, identifying the best virtual machine configuration in terms of the optimal processor allocation for executing parallel and distributed applications. Through the study, it was achieved ~3 times improvements of the elapse time on scientific and transactional applications.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133899521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rule-based classifiers have been successfully applied in data mining applications. In this Paper, we have proposed a novel rule generator classifier called CORER (Colonial competitive Rule-based classifier) to improve the accuracy of data classification. The proposed classifier works based on CCA (Colonial Competitive Algorithm), a recently-developed evolutionary optimization algorithm. In order to approve the CORER capability in various domains, four different datasets from UCI machine learning database repository have been applied. To evaluate CORER performance, we compared our results with some other well-known classification methods, such as C4.5, CN.2, ID3 and naïve bayes which brings about superior results. Our findings lead us to believe that CORER may provide better performance for some critic domains which need more precise classifiers.
{"title":"CORER: A New Rule Generator Classifier","authors":"Javad Basiri, F. Taghiyareh, Sahar Gazani","doi":"10.1109/CSE.2010.18","DOIUrl":"https://doi.org/10.1109/CSE.2010.18","url":null,"abstract":"Rule-based classifiers have been successfully applied in data mining applications. In this Paper, we have proposed a novel rule generator classifier called CORER (Colonial competitive Rule-based classifier) to improve the accuracy of data classification. The proposed classifier works based on CCA (Colonial Competitive Algorithm), a recently-developed evolutionary optimization algorithm. In order to approve the CORER capability in various domains, four different datasets from UCI machine learning database repository have been applied. To evaluate CORER performance, we compared our results with some other well-known classification methods, such as C4.5, CN.2, ID3 and naïve bayes which brings about superior results. Our findings lead us to believe that CORER may provide better performance for some critic domains which need more precise classifiers.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"27 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132604930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Liang Yuan, Yunquan Zhang, Yuxin Tang, L. Rao, Xiangzheng Sun
In large-scale cluster systems, interconnecting thousands of computing nodes increase the complexity of the network topology. Nevertheless, few existing computational models consider the impact of hierarchical communication latencies and bandwidths caused by the network complexity. In this paper we propose a new parallel computational model called LogGPH with a new parameter H incorporated into the LogGP model to describe the communication hierarchy. Through predicting and analyzing the point-to-point and collective MPI_Allgather communication on two 100-Terascale supercomputers, the Dawning 5000A and the Deep Comp 7000, with the new model, it shows that the new model is more accurate than the LogGP model. The mean of absolute error of our model on point-to-point communications is 13%, but the value is 30% without the hierarchical communication consideration.
{"title":"LogGPH: A Parallel Computational Model with Hierarchical Communication Awareness","authors":"Liang Yuan, Yunquan Zhang, Yuxin Tang, L. Rao, Xiangzheng Sun","doi":"10.1109/CSE.2010.40","DOIUrl":"https://doi.org/10.1109/CSE.2010.40","url":null,"abstract":"In large-scale cluster systems, interconnecting thousands of computing nodes increase the complexity of the network topology. Nevertheless, few existing computational models consider the impact of hierarchical communication latencies and bandwidths caused by the network complexity. In this paper we propose a new parallel computational model called LogGPH with a new parameter H incorporated into the LogGP model to describe the communication hierarchy. Through predicting and analyzing the point-to-point and collective MPI_Allgather communication on two 100-Terascale supercomputers, the Dawning 5000A and the Deep Comp 7000, with the new model, it shows that the new model is more accurate than the LogGP model. The mean of absolute error of our model on point-to-point communications is 13%, but the value is 30% without the hierarchical communication consideration.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115923194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Automatically classifying academic conference into semantic topic promises improved academic search and browsing for users. Social tagging is an increasingly popular way of describing the topic of academic conference. However, no attention has been devoted to academic conference classification by making use of social tags. Motivated by this observation, this paper proposes a method which utilizes social tags as well as the content of academic conference in order to improve automatically identifying academic conference classification. The proposed method applies different automatic classification algorithms to improve classification quality by using social tags. Experimental results show that this method mentioned above performs better than the method which only utilizes the content to classify academic conference with 1% Precision measure score increase and 1.64% F1 measure score increase, which demonstrates the effectiveness of the proposed method.
{"title":"Optimizing Academic Conference Classification Using Social Tags","authors":"Jing Xia, Kunmei Wen, Ruixuan Li, X. Gu","doi":"10.1109/CSE.2010.43","DOIUrl":"https://doi.org/10.1109/CSE.2010.43","url":null,"abstract":"Automatically classifying academic conference into semantic topic promises improved academic search and browsing for users. Social tagging is an increasingly popular way of describing the topic of academic conference. However, no attention has been devoted to academic conference classification by making use of social tags. Motivated by this observation, this paper proposes a method which utilizes social tags as well as the content of academic conference in order to improve automatically identifying academic conference classification. The proposed method applies different automatic classification algorithms to improve classification quality by using social tags. Experimental results show that this method mentioned above performs better than the method which only utilizes the content to classify academic conference with 1% Precision measure score increase and 1.64% F1 measure score increase, which demonstrates the effectiveness of the proposed method.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114903322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Service selection has become a key step for cross-organizational collaboration in service-oriented practices and gained ever-increasing attention in both academic and industrial domains. However, the organizations involved may hold different types of evaluation scores, e.g., crisp number, value range and fuzzy linguistic terms. Besides, for the scores of fuzzy linguistic terms, different organizations may hold various evaluation granularities to meet their personalized preferences, which further increase the difficulties for unified service evaluation. So it is a great challenge to take these aspects into consideration for cross-organizational service selection. In view of this challenge, a comprehensive evaluation method named CRML (Crisp number-value Range-Multiple granularities Linguistic terms, CRML) is put forward in this paper. First, the scores of various evaluation types are unified into a form of trapezoidal fuzzy numbers. Second, a classic TOPSIS method is employed to rank all the candidate services for cross-organizational service selection. Finally, a case study is brought forth to validate the feasibility of our proposal.
{"title":"A Comprehensive Evaluation Method for Cross-Organizational Service Selection","authors":"Rutao Yang, Lianyong Qi, Wenmin Lin, Wanchun Dou, Jinjun Chen","doi":"10.1109/CSE.2010.54","DOIUrl":"https://doi.org/10.1109/CSE.2010.54","url":null,"abstract":"Service selection has become a key step for cross-organizational collaboration in service-oriented practices and gained ever-increasing attention in both academic and industrial domains. However, the organizations involved may hold different types of evaluation scores, e.g., crisp number, value range and fuzzy linguistic terms. Besides, for the scores of fuzzy linguistic terms, different organizations may hold various evaluation granularities to meet their personalized preferences, which further increase the difficulties for unified service evaluation. So it is a great challenge to take these aspects into consideration for cross-organizational service selection. In view of this challenge, a comprehensive evaluation method named CRML (Crisp number-value Range-Multiple granularities Linguistic terms, CRML) is put forward in this paper. First, the scores of various evaluation types are unified into a form of trapezoidal fuzzy numbers. Second, a classic TOPSIS method is employed to rank all the candidate services for cross-organizational service selection. Finally, a case study is brought forth to validate the feasibility of our proposal.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"204 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121235338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We consider the system-level self-diagnosis of multiprocessor and multicomputer systems under the generalized comparison model (GCM). In this diagnosis model, a set of tasks is assigned to pairs of nodes and their outcomes are compared by neighboring nodes. The collections of all comparison outcomes, agreements and disagreements among the nodes, are used to identify the set of faulty nodes. We consider only permanent faults in t-diagnosable systems that guarantee that each node can be correctly identified as fault-free or faulty based on a valid collection of comparison results (the syndrome) and assuming that the number of faulty nodes does not exceed a given bound t. Given that comparisons are performed by the nodes themselves, faulty nodes can incorrectly claim that fault-free nodes are faulty or that faulty nodes are fault-free. In this paper, we introduce a novel neural networks-based diagnosis approach to solve this fault identification problem. The new diagnosis approach exploits the off-line learning phase of neural networks to speed up the diagnosis algorithm. We have implemented and evaluated the new diagnosis approach using randomly generated diagnosable systems. The new neural-network-based self-diagnosis approach correctly identified most of the faulty situations forming hence a viable addition or alternative to solve the GCM-based fault identification problem.
{"title":"A Novel Generalized-Comparison-Based Self-Diagnosis Algorithm for Multiprocessor and Multicomputer Systems Using a Multilayered Neural Network","authors":"M. Elhadef, A. Nayak","doi":"10.1109/CSE.2010.68","DOIUrl":"https://doi.org/10.1109/CSE.2010.68","url":null,"abstract":"We consider the system-level self-diagnosis of multiprocessor and multicomputer systems under the generalized comparison model (GCM). In this diagnosis model, a set of tasks is assigned to pairs of nodes and their outcomes are compared by neighboring nodes. The collections of all comparison outcomes, agreements and disagreements among the nodes, are used to identify the set of faulty nodes. We consider only permanent faults in t-diagnosable systems that guarantee that each node can be correctly identified as fault-free or faulty based on a valid collection of comparison results (the syndrome) and assuming that the number of faulty nodes does not exceed a given bound t. Given that comparisons are performed by the nodes themselves, faulty nodes can incorrectly claim that fault-free nodes are faulty or that faulty nodes are fault-free. In this paper, we introduce a novel neural networks-based diagnosis approach to solve this fault identification problem. The new diagnosis approach exploits the off-line learning phase of neural networks to speed up the diagnosis algorithm. We have implemented and evaluated the new diagnosis approach using randomly generated diagnosable systems. The new neural-network-based self-diagnosis approach correctly identified most of the faulty situations forming hence a viable addition or alternative to solve the GCM-based fault identification problem.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130283161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Next-generation, high-throughput sequencers are now capable of producing hundreds of billions of short sequences (reads) in a single day. The task of accurately mapping the reads back to a reference genome is of particular importance because it is used in several other biological applications, e.g., genome re-sequencing, DNA methylation, and ChiP sequencing. On a personal computer (PC), the computationally intensive short-read mapping task currently requires several hours to execute while working on very large sets of reads and genomes. Accelerating this task requires parallel computing. Among the current parallel computing platforms, the graphics processing unit (GPU) provides massively parallel computational prowess that holds the promise of accelerating scientific applications at low cost. In this paper, we propose GPU-RMAP, a massively parallel version of the RMAP short-read mapping tool that is highly optimized for the NVIDIA family of GPUs. We then evaluate GPU-RMAP by mapping millions of synthetic and real reads of varying widths on the mosquito (Aedes aegypti) and human genomes. We also discuss the effects of various input parameters, such as read width, number of reads, and chromosome size, on the performance of GPU-RMAP. We then show that despite using the conventionally “slower” but GPU-compatible binary search algorithm, GPU-RMAP outperforms the sequential RMAP implementation, which uses the “faster” hashing technique on a PC. Our data-parallel GPU implementation results in impressive speedups of up to 14:5-times for the mapping kernel and up to 9:6-times for the overall program execution time over the sequential RMAP implementation on a traditional PC.
{"title":"GPU-RMAP: Accelerating Short-Read Mapping on Graphics Processors","authors":"Ashwin M. Aji, Liqing Zhang, Wu-chun Feng","doi":"10.1109/CSE.2010.29","DOIUrl":"https://doi.org/10.1109/CSE.2010.29","url":null,"abstract":"Next-generation, high-throughput sequencers are now capable of producing hundreds of billions of short sequences (reads) in a single day. The task of accurately mapping the reads back to a reference genome is of particular importance because it is used in several other biological applications, e.g., genome re-sequencing, DNA methylation, and ChiP sequencing. On a personal computer (PC), the computationally intensive short-read mapping task currently requires several hours to execute while working on very large sets of reads and genomes. Accelerating this task requires parallel computing. Among the current parallel computing platforms, the graphics processing unit (GPU) provides massively parallel computational prowess that holds the promise of accelerating scientific applications at low cost. In this paper, we propose GPU-RMAP, a massively parallel version of the RMAP short-read mapping tool that is highly optimized for the NVIDIA family of GPUs. We then evaluate GPU-RMAP by mapping millions of synthetic and real reads of varying widths on the mosquito (Aedes aegypti) and human genomes. We also discuss the effects of various input parameters, such as read width, number of reads, and chromosome size, on the performance of GPU-RMAP. We then show that despite using the conventionally “slower” but GPU-compatible binary search algorithm, GPU-RMAP outperforms the sequential RMAP implementation, which uses the “faster” hashing technique on a PC. Our data-parallel GPU implementation results in impressive speedups of up to 14:5-times for the mapping kernel and up to 9:6-times for the overall program execution time over the sequential RMAP implementation on a traditional PC.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114330617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Augmented Reality allows the user to see the virtual objects superimposed upon or composited with the real world. This paper presented the system design of the Multi-Object Oriented Augmented Reality (MOOAR) system for location-based adaptive mobile learning environment and the scenario study. Moreover, the detailed rationales behind the MOOAR system are also discussed in this paper. The implementation of the MOOAR system is described with the designed scenario. Furthermore, the expected results of the scenario study are shown in this paper to demonstrate the advantages of using Augmented Reality in location-based adaptive mobile learning.
{"title":"Augmented Reality System Design and Scenario Study for Location-Based Adaptive Mobile Learning","authors":"William Chang, Qing Tan","doi":"10.1109/CSE.2010.66","DOIUrl":"https://doi.org/10.1109/CSE.2010.66","url":null,"abstract":"Augmented Reality allows the user to see the virtual objects superimposed upon or composited with the real world. This paper presented the system design of the Multi-Object Oriented Augmented Reality (MOOAR) system for location-based adaptive mobile learning environment and the scenario study. Moreover, the detailed rationales behind the MOOAR system are also discussed in this paper. The implementation of the MOOAR system is described with the designed scenario. Furthermore, the expected results of the scenario study are shown in this paper to demonstrate the advantages of using Augmented Reality in location-based adaptive mobile learning.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128720458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Among gestures in non-verbal communication, pointing gesture can be taken as one of natural human computer interfaces. Vision based hand pointing is an optimal model for human-computer interaction (HCI). One of key problems among the vision based pointing gesture is how to recognize the pointing. Aiming at some limits existing in the literature, a novel method is developed to estimate pointing gestures based on some non-calibrated cameras. Multiple un-calibrated cameras are adopted to determine the pointing target based on pointing features extracted from multiple cameras and support vector machine (SVM) classifier. No explicit constraints are set on the cameras placement. Pointing user can move freely inside a wider interaction environment while pointing at some targets. The mentioned approach does not constrain the pointing surface whether is flat or not, or the target is visible by the cameras. Edge detection based on multi-scale wavelet transformation is used to extract pointing objects from a clutter background. Experiments have shown that the developed approach is efficient for pointing recognition by comparisons.
{"title":"Uncalibrated Camera Vision Pointing Recognition for HCI","authors":"Ye-peng Guan","doi":"10.1109/CSE.2010.34","DOIUrl":"https://doi.org/10.1109/CSE.2010.34","url":null,"abstract":"Among gestures in non-verbal communication, pointing gesture can be taken as one of natural human computer interfaces. Vision based hand pointing is an optimal model for human-computer interaction (HCI). One of key problems among the vision based pointing gesture is how to recognize the pointing. Aiming at some limits existing in the literature, a novel method is developed to estimate pointing gestures based on some non-calibrated cameras. Multiple un-calibrated cameras are adopted to determine the pointing target based on pointing features extracted from multiple cameras and support vector machine (SVM) classifier. No explicit constraints are set on the cameras placement. Pointing user can move freely inside a wider interaction environment while pointing at some targets. The mentioned approach does not constrain the pointing surface whether is flat or not, or the target is visible by the cameras. Edge detection based on multi-scale wavelet transformation is used to extract pointing objects from a clutter background. Experiments have shown that the developed approach is efficient for pointing recognition by comparisons.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"306 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116797842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In a virtualized environment, all the resource are managed by a virtual machine monitor (VMM). Virtualization technology creates separate, isolated and secure running environments for operating systems, programs or applications respectively. As the rapid development of hardware, the computing resource becomes abundant, which is a great chance for virtualization to extend its utilization, besides the amount of virtual machines (VMs) built and supported on a solid computing platform also increases rapidly. When a large amount of communication intensive software, such as web service, database center and gateway or domain name server, are deploying on virtualized environment, they have a demand for communicate with each other. For a mature virtual machine monitor, LAN-styled network communication mechanism for co-located virtual machines is a component of necessity. Although this is easy to use and transparent to user space programs and applications, the performance is often not so good to meet this demand because of the isolation barrier brought by the VMM. In this paper, we propose ColorCom2, a transparent co-located virtual machine communication mechanism. It applies directly memory copying and bypasses the traditional split driver model, producing a high performance in co-located virtual machine communication, and it also keeps transparent to upper level programs so any modification of program is not necessary. Also, ColorCom2 has an advantage over some other similar work is that it can work well although in the case that the underlay network device is interrupted. We use benchmarks and common programs to testColorCom2 in Xen hyper visor and the evaluation result have demonstrated that it has an explicit performance boost and a lower resource cost than the in-built co-located virtual machine communication mechanism. Meanwhile, the philosophy behind the design and implementation of ColorCom2 is almost universally applicable in any type of virtual machine.
{"title":"ColorCom2: A Transparent Co-located Virtual Machine Communication Mechanism","authors":"L. Zhang, Yuebin Bai, Ming Liu, Hanwen Xu","doi":"10.1109/CSE.2010.19","DOIUrl":"https://doi.org/10.1109/CSE.2010.19","url":null,"abstract":"In a virtualized environment, all the resource are managed by a virtual machine monitor (VMM). Virtualization technology creates separate, isolated and secure running environments for operating systems, programs or applications respectively. As the rapid development of hardware, the computing resource becomes abundant, which is a great chance for virtualization to extend its utilization, besides the amount of virtual machines (VMs) built and supported on a solid computing platform also increases rapidly. When a large amount of communication intensive software, such as web service, database center and gateway or domain name server, are deploying on virtualized environment, they have a demand for communicate with each other. For a mature virtual machine monitor, LAN-styled network communication mechanism for co-located virtual machines is a component of necessity. Although this is easy to use and transparent to user space programs and applications, the performance is often not so good to meet this demand because of the isolation barrier brought by the VMM. In this paper, we propose ColorCom2, a transparent co-located virtual machine communication mechanism. It applies directly memory copying and bypasses the traditional split driver model, producing a high performance in co-located virtual machine communication, and it also keeps transparent to upper level programs so any modification of program is not necessary. Also, ColorCom2 has an advantage over some other similar work is that it can work well although in the case that the underlay network device is interrupted. We use benchmarks and common programs to testColorCom2 in Xen hyper visor and the evaluation result have demonstrated that it has an explicit performance boost and a lower resource cost than the in-built co-located virtual machine communication mechanism. Meanwhile, the philosophy behind the design and implementation of ColorCom2 is almost universally applicable in any type of virtual machine.","PeriodicalId":342688,"journal":{"name":"2010 13th IEEE International Conference on Computational Science and Engineering","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115464515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：