In this paper, we describe disparity, a tool that does parallel, scalable anomaly detection for clusters. Disparity uses basic statistical methods and scalable reduction operations to perform data reduction on client nodes and uses these results to locate node anomalies. We discuss the implementation of disparity and present results of its use on a SiCortex SC5832 system.
{"title":"Disparity: Scalable Anomaly Detection for Clusters","authors":"N. Desai, Rick Bradshaw, E. Lusk","doi":"10.1109/ICPP-W.2008.30","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.30","url":null,"abstract":"In this paper, we describe disparity, a tool that does parallel, scalable anomaly detection for clusters. Disparity uses basic statistical methods and scalable reduction operations to perform data reduction on client nodes and uses these results to locate node anomalies. We discuss the implementation of disparity and present results of its use on a SiCortex SC5832 system.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129488338","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 address the issue of the development of performance models for programs that may be executed on large-scale computing systems. The commonly used approaches apply non-standard notations for model specification and often require that the software engineer has a thorough understanding of the underlying performance modeling technique. We propose to bridge the gap between the performance modeling and software engineering by incorporating UML. In our approach we aim to permit the graphical specification of performance model in a human-intuitive fashion on one hand, but on the other hand we aim for a machine-efficient model evaluation. The user specifies graphically the performance model using UML. Thereafter, the transformation of the performance model from the human-usable UML representation to the machine-efficient C++ representation is done automatically. We describe our methodology and illustrate it with the automatic transformation of a sample performance model.
{"title":"Automatic Performance Model Transformation from UML to C++","authors":"Sabri Pllana, S. Benkner, F. Xhafa, L. Barolli","doi":"10.1109/ICPP-W.2008.29","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.29","url":null,"abstract":"We address the issue of the development of performance models for programs that may be executed on large-scale computing systems. The commonly used approaches apply non-standard notations for model specification and often require that the software engineer has a thorough understanding of the underlying performance modeling technique. We propose to bridge the gap between the performance modeling and software engineering by incorporating UML. In our approach we aim to permit the graphical specification of performance model in a human-intuitive fashion on one hand, but on the other hand we aim for a machine-efficient model evaluation. The user specifies graphically the performance model using UML. Thereafter, the transformation of the performance model from the human-usable UML representation to the machine-efficient C++ representation is done automatically. We describe our methodology and illustrate it with the automatic transformation of a sample performance model.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115099180","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}
Load balancing is performed to achieve the optimal use of the existing computational resources as much as possible whereby none of the resources remains idle while some other resources are being utilized. Balanced load distribution can be achieved by the immigration of the load from the source nodes which have surplus workload to the comparatively lightly loaded destination nodes. Applying load balancing during run time is called dynamic load balancing (DLB). DLB can be realized both in a direct or iterative manner according to the execution node selection. In iterative methods, the final destination node is determined through several iteration steps, while in direct methods it is selected in one step. This paper presents the randomized hydrodynamic load balancing (RHLB) method which is a hybrid method that takes advantage of both direct and iterative methods. Using random load migration as a direct method, RHLB approach intends to solve the problems derived from the exceptional instantaneous load rises, and diffuse the surplus workload to relatively free resources. Besides, using hydrodynamic approach as an iterative method, RHLB aims to consume minimum possible system resources to balance the common workload distributions. The results of the experiments designate that, RHLB outruns other iterative based methods in terms of both balance quality and the total time of the load balancing process.
{"title":"Randomized Hydrodynamic Load Balancing Approach","authors":"O. Destanoglu, F. E. Sevilgen","doi":"10.1109/ICPP-W.2008.8","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.8","url":null,"abstract":"Load balancing is performed to achieve the optimal use of the existing computational resources as much as possible whereby none of the resources remains idle while some other resources are being utilized. Balanced load distribution can be achieved by the immigration of the load from the source nodes which have surplus workload to the comparatively lightly loaded destination nodes. Applying load balancing during run time is called dynamic load balancing (DLB). DLB can be realized both in a direct or iterative manner according to the execution node selection. In iterative methods, the final destination node is determined through several iteration steps, while in direct methods it is selected in one step. This paper presents the randomized hydrodynamic load balancing (RHLB) method which is a hybrid method that takes advantage of both direct and iterative methods. Using random load migration as a direct method, RHLB approach intends to solve the problems derived from the exceptional instantaneous load rises, and diffuse the surplus workload to relatively free resources. Besides, using hydrodynamic approach as an iterative method, RHLB aims to consume minimum possible system resources to balance the common workload distributions. The results of the experiments designate that, RHLB outruns other iterative based methods in terms of both balance quality and the total time of the load balancing process.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125948571","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 investigated the problem of supertasking in Pfair-scheduled multi-processor systems. In this approach, a set of tasks (component tasks) is grouped together to form a supertask, which is then scheduled as an ordinary Pfair task. Whenever a supertask is scheduled, its processor time is allocated to its component tasks according to an internal scheduling algorithm. Supertasking approach does not provide guarantees that its component tasks will respect the timeline constraints. In this paper, we propose a new condition for constructing a supertask from its component tasks in such a way that all timing constraints of its component tasks are guaranteed without compromising on schedulability loss. This condition is expressed as weight bound relation b/w supertask and its component tasks.
{"title":"Weight Bound Limits in Supertasking Approach for Guaranteed Timeline Constraints","authors":"Muhammad Farooq, F. Muller, M. Auguin","doi":"10.1109/ICPP-W.2008.24","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.24","url":null,"abstract":"We investigated the problem of supertasking in Pfair-scheduled multi-processor systems. In this approach, a set of tasks (component tasks) is grouped together to form a supertask, which is then scheduled as an ordinary Pfair task. Whenever a supertask is scheduled, its processor time is allocated to its component tasks according to an internal scheduling algorithm. Supertasking approach does not provide guarantees that its component tasks will respect the timeline constraints. In this paper, we propose a new condition for constructing a supertask from its component tasks in such a way that all timing constraints of its component tasks are guaranteed without compromising on schedulability loss. This condition is expressed as weight bound relation b/w supertask and its component tasks.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121088670","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 wireless sensor-actuator network (WSAN), sensor nodes deliver messages with sensed values to only nearby nodes due to weak radio. Messages are forwarded by sensor nodes to an actuator node by a routing protocol like a flooding protocol. In the redundant data transmission (RT) protocol, a sensor node sends a message with not only its sensed value but also sensed values received from other sensor nodes. Each message carries so many number of sensed values that the message loss ratio is not increased. Even if a message with a sensed value v is lost, an actuator node can receive the value v from a message sent by another sensor node. Thus, each sensed value is redundantly carried in multiple messages. The redundancy of a sensed value is in nature increased by broadcasting the value. In order to reduce the redundancy, we take a strategy that the farther sensor nodes from an actuator node forward the fewer number of sensed values. We evaluate the RT protocol in terms of loss ratio, redundancy, and delay time of sensed values. We show that about 80% of sensed values can be delivered to an actuator node even if 95% of messages are lost due to noise and collision.
{"title":"A Redundant Data Transmission Protocol for a Wireless Sensor Network","authors":"A. Aikebaier, T. Enokido, M. Takizawa","doi":"10.1109/ICPP-W.2008.12","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.12","url":null,"abstract":"In a wireless sensor-actuator network (WSAN), sensor nodes deliver messages with sensed values to only nearby nodes due to weak radio. Messages are forwarded by sensor nodes to an actuator node by a routing protocol like a flooding protocol. In the redundant data transmission (RT) protocol, a sensor node sends a message with not only its sensed value but also sensed values received from other sensor nodes. Each message carries so many number of sensed values that the message loss ratio is not increased. Even if a message with a sensed value v is lost, an actuator node can receive the value v from a message sent by another sensor node. Thus, each sensed value is redundantly carried in multiple messages. The redundancy of a sensed value is in nature increased by broadcasting the value. In order to reduce the redundancy, we take a strategy that the farther sensor nodes from an actuator node forward the fewer number of sensed values. We evaluate the RT protocol in terms of loss ratio, redundancy, and delay time of sensed values. We show that about 80% of sensed values can be delivered to an actuator node even if 95% of messages are lost due to noise and collision.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130371059","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}
Platform and resource virtualization is evolving as an enabling technology to consolidate servers and to simplify the management of enterprise servers. Provisioning of virtual machines (VM) on demand is a key component of such environments. This paper presents Iceberg, a space and time efficient image streamer for provisioning VMs. Iceberg operates at block storage (image) leveland is OS, Hypervisor, and file system agnostic. Iceberg is a chunk-based distributed image streamer which provides instant provisioning of VMs over the network. Iceberg utilizes a variable-size, content-based slicing algorithm such that the sharing among images is improved and storage space requirement is reduced significantly. We provide a comprehensive study of sharing among several different types of images and show up to 60% content sharing between two images. Much higher storage space saving can be achieved among a large number of similar images.
{"title":"Iceberg: An Image Streamer for Space and Time Efficient Provisioning of Virtual Machines","authors":"Lei Shi, M. Banikazemi, Qingqing Wang","doi":"10.1109/ICPP-W.2008.13","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.13","url":null,"abstract":"Platform and resource virtualization is evolving as an enabling technology to consolidate servers and to simplify the management of enterprise servers. Provisioning of virtual machines (VM) on demand is a key component of such environments. This paper presents Iceberg, a space and time efficient image streamer for provisioning VMs. Iceberg operates at block storage (image) leveland is OS, Hypervisor, and file system agnostic. Iceberg is a chunk-based distributed image streamer which provides instant provisioning of VMs over the network. Iceberg utilizes a variable-size, content-based slicing algorithm such that the sharing among images is improved and storage space requirement is reduced significantly. We provide a comprehensive study of sharing among several different types of images and show up to 60% content sharing between two images. Much higher storage space saving can be achieved among a large number of similar images.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115273203","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}
Abhijit C. Navalekar, Jitish S. Kolanjery, W. Michalson, J. Matthews
This paper investigates the throughput performance of a CSMA/CA based network of land mobile radios (LMR) over a Rayleigh fading FM channel. The throughput calculations are based on the medium access (MAC)/physical (PHY) cross-layer characteristics for the distributed digital radios (DDR). From PHY-layer perspective, the effects of additive white Gaussian noise (AWGN), Rayleigh fading and push-to-talk (PTT) delays are considered while the MAC-layer perspective includes the performance of carrier sense multiple access with collision avoidance (CSMA/CA) for a half duplex narrowband FM (NBFM) channel. The throughput performance was simulated for channel data rates of 12000 bps in presence of additive white Gaussian noise (AWGN) and Rayleigh fading. Typical values for PTT delays for two of the most commonly used LMR terminals were calculated experimentally. The results show that in addition to fading and number of DDR nodes in a network, the push-to-talk (PTT) delays associated with LMR radios significantly influence the achievable throughput of the system.
{"title":"Effects of Push-to-Talk (PTT) Delays on Throughput Performance of CSMA/CA Based Distributed Digital Radios (DDR) for Land Mobile Radio (LMR) Networks","authors":"Abhijit C. Navalekar, Jitish S. Kolanjery, W. Michalson, J. Matthews","doi":"10.1109/ICPP-W.2008.14","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.14","url":null,"abstract":"This paper investigates the throughput performance of a CSMA/CA based network of land mobile radios (LMR) over a Rayleigh fading FM channel. The throughput calculations are based on the medium access (MAC)/physical (PHY) cross-layer characteristics for the distributed digital radios (DDR). From PHY-layer perspective, the effects of additive white Gaussian noise (AWGN), Rayleigh fading and push-to-talk (PTT) delays are considered while the MAC-layer perspective includes the performance of carrier sense multiple access with collision avoidance (CSMA/CA) for a half duplex narrowband FM (NBFM) channel. The throughput performance was simulated for channel data rates of 12000 bps in presence of additive white Gaussian noise (AWGN) and Rayleigh fading. Typical values for PTT delays for two of the most commonly used LMR terminals were calculated experimentally. The results show that in addition to fading and number of DDR nodes in a network, the push-to-talk (PTT) delays associated with LMR radios significantly influence the achievable throughput of the system.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124806731","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}
Workflow applications are gaining popularity in recent years because of the prevalence of cluster environments. Many algorithms have been developed since, however most static algorithms are designed in the problem domain of scheduling single workflow applications, thus not applicable to a common cluster environment where multiple workflow applications and other independent jobs compete for resources. Dynamic scheduling approaches can handle the mixed workload practically by nature but their performance has yet to optimize as they do not have a global view of workflow applications. Recent research efforts suggest merging multiple workflows into one workflow before execution, but fail to address an important issue that multiple workflow applications may be submitted at different times by different users. In this paper, we propose a planner-guided dynamic scheduling strategy for multiple workflow applications, leveraging job dependence information and execution time estimation.Our approach schedules individual jobs dynamically without requiring merging the workflow applications a priori. The simulation results show that the proposed algorithm significantly outperforms two other algorithms by 43.6% and 36.7% with respect to workflow makespan and turnaround time respectively, and it performs even better when the number of concurrent workflow applications increases and the resources are scarce.
{"title":"A Planner-Guided Scheduling Strategy for Multiple Workflow Applications","authors":"Zhifeng Yu, Weisong Shi","doi":"10.1109/ICPP-W.2008.10","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.10","url":null,"abstract":"Workflow applications are gaining popularity in recent years because of the prevalence of cluster environments. Many algorithms have been developed since, however most static algorithms are designed in the problem domain of scheduling single workflow applications, thus not applicable to a common cluster environment where multiple workflow applications and other independent jobs compete for resources. Dynamic scheduling approaches can handle the mixed workload practically by nature but their performance has yet to optimize as they do not have a global view of workflow applications. Recent research efforts suggest merging multiple workflows into one workflow before execution, but fail to address an important issue that multiple workflow applications may be submitted at different times by different users. In this paper, we propose a planner-guided dynamic scheduling strategy for multiple workflow applications, leveraging job dependence information and execution time estimation.Our approach schedules individual jobs dynamically without requiring merging the workflow applications a priori. The simulation results show that the proposed algorithm significantly outperforms two other algorithms by 43.6% and 36.7% with respect to workflow makespan and turnaround time respectively, and it performs even better when the number of concurrent workflow applications increases and the resources are scarce.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"30 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116337578","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}
Anne-Elisabeth Baert, Vincent Boudet, A. Jean-Marie, Xavier Roche
In this paper, we examine the problem of minimizing the variance of the download time in a particular Video on Demand System. This VOD system is based on a Grid Delivery Network which is an hybrid architecture based on P2P and Grid Computing concepts. In this system, videos are divided into blocks and replicated on hosts to decrease the average response time. The purpose of the paper is to study the impact of the block allocation scheme on the variance of the download time. We formulate this as an optimization problem, and show that this problem can be reduced to finding a Steiner System. We analyze different heuristics to solve it in practice, and validate through simulation that a random allocation is quasi-optimal.
{"title":"Optimization of Download Times in a Distributed VOD System","authors":"Anne-Elisabeth Baert, Vincent Boudet, A. Jean-Marie, Xavier Roche","doi":"10.1109/ICPP-W.2008.27","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.27","url":null,"abstract":"In this paper, we examine the problem of minimizing the variance of the download time in a particular Video on Demand System. This VOD system is based on a Grid Delivery Network which is an hybrid architecture based on P2P and Grid Computing concepts. In this system, videos are divided into blocks and replicated on hosts to decrease the average response time. The purpose of the paper is to study the impact of the block allocation scheme on the variance of the download time. We formulate this as an optimization problem, and show that this problem can be reduced to finding a Steiner System. We analyze different heuristics to solve it in practice, and validate through simulation that a random allocation is quasi-optimal.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121448240","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}
The use of discrete-event simulators in the design and development of distributed systems is appealing due to their efficiency and scalability. Their core abstractions of process and event map neatly to the components and interactions of modern-day distributed systems and allow designing realistic simulation scenarios. MONARC, a multi-threaded, process oriented simulation framework designed for modeling large scale distributed systems, allows the realistic simulation of a wide-range of distributed system technologies, with respect to their specific components and characteristics. In this paper we present an innovative solution to the problem of evaluating the dependability characteristic of distributed systems. Our solution is based on several proposed extensions to the simulation model of the MONARC simulation framework. These extensions refer to fault tolerance and system orchestration mechanisms being added in order to assess the reliability and availability of distributed systems. The extended simulation model includes the necessary components to describe various actual failure situations and provides the mechanisms to evaluate different strategies for replication and redundancy procedures, as well as security enforcement mechanisms.
{"title":"A Simulation Framework for Dependable Distributed Systems","authors":"C. Dobre, Florin Pop, V. Cristea","doi":"10.1109/ICPP-W.2008.19","DOIUrl":"https://doi.org/10.1109/ICPP-W.2008.19","url":null,"abstract":"The use of discrete-event simulators in the design and development of distributed systems is appealing due to their efficiency and scalability. Their core abstractions of process and event map neatly to the components and interactions of modern-day distributed systems and allow designing realistic simulation scenarios. MONARC, a multi-threaded, process oriented simulation framework designed for modeling large scale distributed systems, allows the realistic simulation of a wide-range of distributed system technologies, with respect to their specific components and characteristics. In this paper we present an innovative solution to the problem of evaluating the dependability characteristic of distributed systems. Our solution is based on several proposed extensions to the simulation model of the MONARC simulation framework. These extensions refer to fault tolerance and system orchestration mechanisms being added in order to assess the reliability and availability of distributed systems. The extended simulation model includes the necessary components to describe various actual failure situations and provides the mechanisms to evaluate different strategies for replication and redundancy procedures, as well as security enforcement mechanisms.","PeriodicalId":231042,"journal":{"name":"2008 International Conference on Parallel Processing - Workshops","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115572057","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：