Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888730
I. Kume, Masahide Nakamura, Naoya Nitta, Etsuya Shibayama
Recently many frameworks are used in software development without proper documentation, and are misused by application developers in calling framework APIs. Debugging a failure caused by a wrong API call is difficult and requires a proper supporting technique. In our preceding study we developed a dynamic analysis technique to detect possibly unexpected side effects that cause failures. In this paper, we introduce a case study to identify a wrong API call using this technique.
{"title":"Toward a dynamic analysis technique to locate framework misuses that cause unexpected side effects","authors":"I. Kume, Masahide Nakamura, Naoya Nitta, Etsuya Shibayama","doi":"10.1109/SNPD.2014.6888730","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888730","url":null,"abstract":"Recently many frameworks are used in software development without proper documentation, and are misused by application developers in calling framework APIs. Debugging a failure caused by a wrong API call is difficult and requires a proper supporting technique. In our preceding study we developed a dynamic analysis technique to detect possibly unexpected side effects that cause failures. In this paper, we introduce a case study to identify a wrong API call using this technique.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134130099","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888697
Dapeng Liu, Shaochun Xu, Zengdi Cui
Big data has been playing a critical role in modern information technology. In some big data management systems, to simplify client design, data requests from clients are even distributed to all nodes and then are routed to the final correct data storage inside the data cluster. After performing analysis on this working mechanism, we point out some design problems and advocate the client-side access partitioning, i.e., clients know precisely which node of the cluster should be accessed for sought information. This approach could provide a fast access. We also implement a first-stage application based on client-side access partitioning for evaluation purpose and the result demonstrates our approach is effective.
{"title":"Using client-side access partitioning for data clustering in big data applications","authors":"Dapeng Liu, Shaochun Xu, Zengdi Cui","doi":"10.1109/SNPD.2014.6888697","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888697","url":null,"abstract":"Big data has been playing a critical role in modern information technology. In some big data management systems, to simplify client design, data requests from clients are even distributed to all nodes and then are routed to the final correct data storage inside the data cluster. After performing analysis on this working mechanism, we point out some design problems and advocate the client-side access partitioning, i.e., clients know precisely which node of the cluster should be accessed for sought information. This approach could provide a fast access. We also implement a first-stage application based on client-side access partitioning for evaluation purpose and the result demonstrates our approach is effective.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134139353","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888729
Masateru Tsunoda, K. Ono
It is important to establish a benchmark of work efficiency for software maintenance and support. For maintenance and support service providers, the benchmarking is the basis for improvement of their work. For the service purchasers, it is useful to check work efficiency of contracted service provider. To establish a benchmark of work efficiency for software development activities, a cross-company dataset is often used. Data points included in it are collected from various organizations. ISBSG (International Software Benchmarking Standards Group) builds the cross-company dataset of software maintenance and support. In the analysis, we found some pitfalls when one analyzes ISBSG software maintenance and support dataset. If the pitfalls are ignored, spurious relationships would be found. In this paper, we showed some pitfalls, and how to avoid it. It would be very useful for researchers, because ISBSG software maintenance and support dataset may be widely used in the near future.
{"title":"Pitfalls of analyzing a cross-company dataset of software maintenance and support","authors":"Masateru Tsunoda, K. Ono","doi":"10.1109/SNPD.2014.6888729","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888729","url":null,"abstract":"It is important to establish a benchmark of work efficiency for software maintenance and support. For maintenance and support service providers, the benchmarking is the basis for improvement of their work. For the service purchasers, it is useful to check work efficiency of contracted service provider. To establish a benchmark of work efficiency for software development activities, a cross-company dataset is often used. Data points included in it are collected from various organizations. ISBSG (International Software Benchmarking Standards Group) builds the cross-company dataset of software maintenance and support. In the analysis, we found some pitfalls when one analyzes ISBSG software maintenance and support dataset. If the pitfalls are ignored, spurious relationships would be found. In this paper, we showed some pitfalls, and how to avoid it. It would be very useful for researchers, because ISBSG software maintenance and support dataset may be widely used in the near future.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"137 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128081302","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888685
Hassan Almari, C. Boughton
Evidence of the relationship between software architecture including it's styles/patterns and quality attributes continues to grow, but largely remains an art rather than a science in terms of being able to predict a relevant architecture from (known) quality attributes. The aim of this paper is to point out those aspects that influence utilization of software architecture artefacts and their evaluation by software developers, and is part of a continuing study the results of which are intended to aid professional software/system developers with their decisions surrounding choice of (concrete) software architectures. The earlier part of the study produced an analysis report based on a survey titled “Questionnaire on matters relating to Software Patterns - 2012”. The survey was issued to software developers possessing more than 5 years experience, and produced significant results which, in turn, led to the need for this second survey. Ninety seven (97%) percent of the participants, from six different nations, answering the first questionnaire supported this further investigation.
{"title":"Questionnaire report on matter relating to software architecture evaluation","authors":"Hassan Almari, C. Boughton","doi":"10.1109/SNPD.2014.6888685","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888685","url":null,"abstract":"Evidence of the relationship between software architecture including it's styles/patterns and quality attributes continues to grow, but largely remains an art rather than a science in terms of being able to predict a relevant architecture from (known) quality attributes. The aim of this paper is to point out those aspects that influence utilization of software architecture artefacts and their evaluation by software developers, and is part of a continuing study the results of which are intended to aid professional software/system developers with their decisions surrounding choice of (concrete) software architectures. The earlier part of the study produced an analysis report based on a survey titled “Questionnaire on matters relating to Software Patterns - 2012”. The survey was issued to software developers possessing more than 5 years experience, and produced significant results which, in turn, led to the need for this second survey. Ninety seven (97%) percent of the participants, from six different nations, answering the first questionnaire supported this further investigation.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125203226","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888735
Kazuyoshi Ichihashi, Hideki Ichihashi
The purpose of this study was to investigate multisensory effect of fragrance. Can olfactory stimulation induce visual, auditory, taste and tactile image? Fifty-eight participants smelled ten different essential oils, including Grapefruit, Neroli, Clarysage, Roman Chamomile, Sandalwood, Marjoram, Damaskrose, Lemongrass, Bergamot, and Frankincense; the preferences and the evoked color, sound, taste and tactile images in each fragrance were assessed. In addition, the close relationships between a favorable fragrance and a bright color image such as orange and yellow, a round shape image, an acidic and sweet taste image, a high pitch sound image and a silky, downy tactile image were identified. We suggested that the olfactory system is highly linked with the preference system. We represented that olfactory system was highly linked with preference system. We considered that olfactory stimulation evokes visual, auditory, taste and tactile images that accompany the activation of the preference system. We considered that multisensory or crossmodal correspondence is very general in human information processing. It is very important to consider crossmodal correspondences to develop more attractive arts and various types of products for humans.
{"title":"Multisensory images evoked by olfactory stimuli: Crossmodal correspondences in human information processing","authors":"Kazuyoshi Ichihashi, Hideki Ichihashi","doi":"10.1109/SNPD.2014.6888735","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888735","url":null,"abstract":"The purpose of this study was to investigate multisensory effect of fragrance. Can olfactory stimulation induce visual, auditory, taste and tactile image? Fifty-eight participants smelled ten different essential oils, including Grapefruit, Neroli, Clarysage, Roman Chamomile, Sandalwood, Marjoram, Damaskrose, Lemongrass, Bergamot, and Frankincense; the preferences and the evoked color, sound, taste and tactile images in each fragrance were assessed. In addition, the close relationships between a favorable fragrance and a bright color image such as orange and yellow, a round shape image, an acidic and sweet taste image, a high pitch sound image and a silky, downy tactile image were identified. We suggested that the olfactory system is highly linked with the preference system. We represented that olfactory system was highly linked with preference system. We considered that olfactory stimulation evokes visual, auditory, taste and tactile images that accompany the activation of the preference system. We considered that multisensory or crossmodal correspondence is very general in human information processing. It is very important to consider crossmodal correspondences to develop more attractive arts and various types of products for humans.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126053104","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888746
Yang Zheng, Lizhi Cai, Shidong Huang, Zhihong Wang
Virtualization technology not only is the basis of Cloud computing technology, but also plays an important role in Cloud Testing. Cloud Testing takes advantage of virtualization technology to generate VM (virtual machine) resources in the physical machine, and adopts the corresponding strategies to schedule the VM resources. VM scheduling strategies have a crucial impact on the overall performance of Cloud Testing. The paper first introduces the scheduling process of VM in Cloud Testing, and divides the common scheduling strategies into three categories: center on saving energy, center on load balancing and center on Qos performance. Then the common VM scheduling strategies in current Cloud Testing environment are analyzed. Finally, their advantages and disadvantages are also discussed.
{"title":"VM scheduling strategies based on artificial intelligence in Cloud Testing","authors":"Yang Zheng, Lizhi Cai, Shidong Huang, Zhihong Wang","doi":"10.1109/SNPD.2014.6888746","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888746","url":null,"abstract":"Virtualization technology not only is the basis of Cloud computing technology, but also plays an important role in Cloud Testing. Cloud Testing takes advantage of virtualization technology to generate VM (virtual machine) resources in the physical machine, and adopts the corresponding strategies to schedule the VM resources. VM scheduling strategies have a crucial impact on the overall performance of Cloud Testing. The paper first introduces the scheduling process of VM in Cloud Testing, and divides the common scheduling strategies into three categories: center on saving energy, center on load balancing and center on Qos performance. Then the common VM scheduling strategies in current Cloud Testing environment are analyzed. Finally, their advantages and disadvantages are also discussed.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132112889","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888676
Naoshi Sakamoto
In order to communicate in cipher over IP multi-cast, each of joining and leaving participants causes renewing keys. Moreover, the number of renewed keys depends on the key management system. LKH, one of the key management systems, uses a tree structure to manage keys to share with participants. Every node of the tree is given a key, and each leaf of the tree is corresponding to a participant. If all members are handled equally, by using a balanced binary tree, the average number of renewed keys per join and leave is estimated at ⌈log2 n ⌉, where n denotes the number of participants. In this study, we introduce a scenario that the key management system can distinguish between inconstant members and stable members, instead of handling members equally. Under this scenario, our system improves the number of renewing keys efficiently by considering another tree structure against the balanced binary tree structure.
{"title":"An efficient structure for LKH key tree on secure multicast communications","authors":"Naoshi Sakamoto","doi":"10.1109/SNPD.2014.6888676","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888676","url":null,"abstract":"In order to communicate in cipher over IP multi-cast, each of joining and leaving participants causes renewing keys. Moreover, the number of renewed keys depends on the key management system. LKH, one of the key management systems, uses a tree structure to manage keys to share with participants. Every node of the tree is given a key, and each leaf of the tree is corresponding to a participant. If all members are handled equally, by using a balanced binary tree, the average number of renewed keys per join and leave is estimated at ⌈log2 n ⌉, where n denotes the number of participants. In this study, we introduce a scenario that the key management system can distinguish between inconstant members and stable members, instead of handling members equally. Under this scenario, our system improves the number of renewing keys efficiently by considering another tree structure against the balanced binary tree structure.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125932597","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888701
Dominic M. Mezzanotte, J. Dehlinger
Existing Enterprise Architecture (EA) frameworks (EAF) focus on the governance and alignment of an enterprise's strategic business plans and operating model with its Information Technology (IT) capabilities. These technically-oriented processes attempt to simplify the verification and validation of design artifacts used during software development. Yet, many EA projects fail mostly for non-technical reasons. The development of an EA can necessitate organizational change which can influence stakeholder behavior in ways that may be detrimental to the EA. An analysis of EAFs finds that they are deficient in incorporating human behavior and that they avoid any of the consequences of human action during EA development. This paper proposes a behavior-driven EA requirements quality management program designed to encourage stakeholder collaboration and participation in EA.
{"title":"Developing and building a quality management system based on stakeholder behavior for enterprise architecture","authors":"Dominic M. Mezzanotte, J. Dehlinger","doi":"10.1109/SNPD.2014.6888701","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888701","url":null,"abstract":"Existing Enterprise Architecture (EA) frameworks (EAF) focus on the governance and alignment of an enterprise's strategic business plans and operating model with its Information Technology (IT) capabilities. These technically-oriented processes attempt to simplify the verification and validation of design artifacts used during software development. Yet, many EA projects fail mostly for non-technical reasons. The development of an EA can necessitate organizational change which can influence stakeholder behavior in ways that may be detrimental to the EA. An analysis of EAFs finds that they are deficient in incorporating human behavior and that they avoid any of the consequences of human action during EA development. This paper proposes a behavior-driven EA requirements quality management program designed to encourage stakeholder collaboration and participation in EA.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127058264","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}
Bug reports play essential roles in many software engineering tasks. Since validity and performance of these tasks definitely rely on the quality of bug reports, accurate information from bug reports is very important. However, as found in previous study, significant numbers of reports classified as bug are not really a bug. Recent studies proposed techniques to automatically classify bug reports into binary classes, yet there is still more to desire. These bug reports can be classified into multiple classes, which could help to identify what these reports are actually about. Moreover, previous study only looks into one possibility of topic modeling, that is, Latent Dirichlet Allocation (LDA). While LDA has its advantage, parameter tuning is required. In this paper, we propose a nonparametric approach to automatically classify bug reports with, another topic modeling method, Hierarchical Dirichlet Process (HDP). The result indicates that our nonparametric approach performance is comparable to the parametric one. We also examine various aspects of LDA to provide more thoroughly understanding of this process.
{"title":"Comparing hierarchical dirichlet process with latent dirichlet allocation in bug report multiclass classification","authors":"Nachai Limsettho, Hideaki Hata, Ken-ichi Matsumoto","doi":"10.1109/SNPD.2014.6888695","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888695","url":null,"abstract":"Bug reports play essential roles in many software engineering tasks. Since validity and performance of these tasks definitely rely on the quality of bug reports, accurate information from bug reports is very important. However, as found in previous study, significant numbers of reports classified as bug are not really a bug. Recent studies proposed techniques to automatically classify bug reports into binary classes, yet there is still more to desire. These bug reports can be classified into multiple classes, which could help to identify what these reports are actually about. Moreover, previous study only looks into one possibility of topic modeling, that is, Latent Dirichlet Allocation (LDA). While LDA has its advantage, parameter tuning is required. In this paper, we propose a nonparametric approach to automatically classify bug reports with, another topic modeling method, Hierarchical Dirichlet Process (HDP). The result indicates that our nonparametric approach performance is comparable to the parametric one. We also examine various aspects of LDA to provide more thoroughly understanding of this process.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131708107","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}
Pub Date : 2014-06-01DOI: 10.1109/SNPD.2014.6888708
Yasong Zheng, Yuanchao Xu, Haibo Meng, Xiaochun Ye, Lingjun Fan, Futao Miao, Dongrui Fan
MapReduce is a popular parallel programming model to program both large scale clusters and shared-memory multicore systems. While one of the major bottlenecks for shared-memory MapReduce is memory allocation. In this paper, we present a Memory Controlling Model (MCM) that can reduce the overhead of memory allocation by reducing the memory consumption. Based on MCM, we extend the MapReduce framework with low memory requirements, called LMMR (Low Memory consuming MapReduce). We have implemented LMMR on top of Phoenix++, an already highly optimized shared-memory MapReduce from Stanford. We evaluate our system on an Intel shared-memory multicore machine with 16 processing threads and compare it with both Phoenix++ and Hadoop. Experiments on three different popular applications show that, compared to Phoenix++, LMMR saves up to 94% memory and results in a speedup ranging from 1.8X to 3.7X. LMMR also is up to 120 times faster than Hadoop.
MapReduce是一种流行的并行编程模型，用于为大规模集群和共享内存多核系统编程。而共享内存MapReduce的主要瓶颈之一是内存分配。本文提出了一种内存控制模型(Memory control Model, MCM)，该模型可以通过减少内存消耗来减少内存分配的开销。在MCM的基础上，我们扩展了低内存需求的MapReduce框架，称为LMMR (low memory consuming MapReduce)。我们已经在菲尼克斯++上实现了LMMR，这是斯坦福大学已经高度优化的共享内存MapReduce。我们在一台英特尔共享内存多核机器上评估了我们的系统，该机器有16个处理线程，并将其与Phoenix++和Hadoop进行了比较。在三种不同的流行应用程序上进行的实验表明，与Phoenix++相比，LMMR节省了高达94%的内存，并实现了1.8到3.7倍的加速。LMMR也比Hadoop快120倍。
{"title":"Optimizing mapreduce with low memory requirements for shared-memory systems","authors":"Yasong Zheng, Yuanchao Xu, Haibo Meng, Xiaochun Ye, Lingjun Fan, Futao Miao, Dongrui Fan","doi":"10.1109/SNPD.2014.6888708","DOIUrl":"https://doi.org/10.1109/SNPD.2014.6888708","url":null,"abstract":"MapReduce is a popular parallel programming model to program both large scale clusters and shared-memory multicore systems. While one of the major bottlenecks for shared-memory MapReduce is memory allocation. In this paper, we present a Memory Controlling Model (MCM) that can reduce the overhead of memory allocation by reducing the memory consumption. Based on MCM, we extend the MapReduce framework with low memory requirements, called LMMR (Low Memory consuming MapReduce). We have implemented LMMR on top of Phoenix++, an already highly optimized shared-memory MapReduce from Stanford. We evaluate our system on an Intel shared-memory multicore machine with 16 processing threads and compare it with both Phoenix++ and Hadoop. Experiments on three different popular applications show that, compared to Phoenix++, LMMR saves up to 94% memory and results in a speedup ranging from 1.8X to 3.7X. LMMR also is up to 120 times faster than Hadoop.","PeriodicalId":272932,"journal":{"name":"15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129648331","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：