The Experts below are selected from a list of 27216 Experts worldwide ranked by ideXlab platform
Toyotaro Suzumura - One of the best experts on this subject based on the ideXlab platform.
-
An X10-Based Distributed Streaming Graph Database Engine
2017 IEEE 24th International Conference on High Performance Computing (HiPC), 2017Co-Authors: Miyuru Dayarathna, Sanath Jayasena, Sathya Bandara, Nandula Jayamaha, Mahen Herath, Achala Madhushan, Toyotaro SuzumuraAbstract:Streaming graph data mining has become a significant issue in high performance graph mining due to the increasing appearance of graph data sets as streams. In this paper we propose Acacia-Stream which is a scalable distributed streaming graph Database Engine developed with X10 programming language. Graph streams are partitioned using a streaming graph partitioner algorithm in Acacia-Stream and streaming graph processing queries are run on the graph streams. The partitioned data sets are persisted on secondary storage across X10 places. We investigate on the use of three different streaming graph partitioner algorithms called hash, Linear Deterministic Greedy, and Fennel algorithms and report their performance. Furthermore, to demonstrate Acacia-Stream's streaming graph processing capabilities we implement streaming triangle counting with Acacia-Stream. We present performance results gathered from Acacia-Stream with different large scale streaming data sets in both horizontal and vertical scalability experiments. Furthermore, we compare streaming graph loading performance of Acacia-Stream with Neo4j and Oracle's PGX graph Database servers. From these experiments we observed that Acacia-Stream's Fennel partitioner based graph uploader can upload a 948MB rmat22 graph in 1283.42 seconds which is 38% faster than PGX graph Database server and 12.8 times faster than Neo4j Database server. Acacia-Stream's Streaming Partitioner's batch size adjustments based optimizations reduced the time used by the network communications almost by half.
-
acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Conference on Cloud Computing, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in High Performance graph mining in recent years. However, the existing Resource Description Framework (RDF) Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL (an RDF query language) processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we designed and implemented a replication based fault tolerance mechanism for Acacia-RDF. We present performance results gathered from Acacia with different scales of LUBM (Lehigh University Benchmark) RDF benchmark data sets. We make a comparison of Acacia-RDF's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported less than ten seconds elapsed times on 16 places for running the first and the third queries of the LUBM benchmark on LUBM 160 universities data set with 3.6 million vertices and 28.5 million edges which was 1.7GB in size. Through this work we describe and demonstrate the use of X10 language for development of scalable RDF graph data management systems.
-
introducing acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Parallel and Distributed Processing Symposium, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in HPC graph mining in recent years. However, the existing RDF Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we demonstrate the implementation of scalable graph algorithms such as Triangle counting with such partitioned data sets. We present performance results gathered from Acacia with different scales of LUBM RDF benchmark data sets and make a comparison of Acacia's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported approximately 2 seconds elapsed time on 4 places for running the first and third queries of the LUBM benchmark on LUBM scale 40 data set. Through this work we introduce the use of X10 language for scalable RDF graph data management.
-
towards scalable distributed graph Database Engine for hybrid clouds
Proceedings of the 5th International Workshop on Data-Intensive Computing in the Clouds, 2014Co-Authors: Miyuru Dayarathna, Toyotaro SuzumuraAbstract:Large graph data management and mining in clouds has become an important issue in recent times. We propose Acacia which is a distributed graph Database Engine for scalable handling of such large graph data. Acacia operates between the boundaries of private and public clouds. Acacia partitions and stores the graph data in the private cloud during its initial deployment. Acacia bursts into the public cloud when the resources of the private cloud are insufficient to maintain its service-level agreements. We implement Acacia using X10 programming language. We describe how Top-K PageRank has been implemented in Acacia. We report preliminary experiment results conducted with Acacia on a small compute cluster. Acacia is able to upload 69 million edges LiveJournal social network data set in about 10 minutes. Furthermore, Acacia calculates the average out degree of vertices of LiveJournal graph in 2 minutes. These results indicate Acacias potential for handling large graphs.
Miyuru Dayarathna - One of the best experts on this subject based on the ideXlab platform.
-
An X10-Based Distributed Streaming Graph Database Engine
2017 IEEE 24th International Conference on High Performance Computing (HiPC), 2017Co-Authors: Miyuru Dayarathna, Sanath Jayasena, Sathya Bandara, Nandula Jayamaha, Mahen Herath, Achala Madhushan, Toyotaro SuzumuraAbstract:Streaming graph data mining has become a significant issue in high performance graph mining due to the increasing appearance of graph data sets as streams. In this paper we propose Acacia-Stream which is a scalable distributed streaming graph Database Engine developed with X10 programming language. Graph streams are partitioned using a streaming graph partitioner algorithm in Acacia-Stream and streaming graph processing queries are run on the graph streams. The partitioned data sets are persisted on secondary storage across X10 places. We investigate on the use of three different streaming graph partitioner algorithms called hash, Linear Deterministic Greedy, and Fennel algorithms and report their performance. Furthermore, to demonstrate Acacia-Stream's streaming graph processing capabilities we implement streaming triangle counting with Acacia-Stream. We present performance results gathered from Acacia-Stream with different large scale streaming data sets in both horizontal and vertical scalability experiments. Furthermore, we compare streaming graph loading performance of Acacia-Stream with Neo4j and Oracle's PGX graph Database servers. From these experiments we observed that Acacia-Stream's Fennel partitioner based graph uploader can upload a 948MB rmat22 graph in 1283.42 seconds which is 38% faster than PGX graph Database server and 12.8 times faster than Neo4j Database server. Acacia-Stream's Streaming Partitioner's batch size adjustments based optimizations reduced the time used by the network communications almost by half.
-
acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Conference on Cloud Computing, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in High Performance graph mining in recent years. However, the existing Resource Description Framework (RDF) Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL (an RDF query language) processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we designed and implemented a replication based fault tolerance mechanism for Acacia-RDF. We present performance results gathered from Acacia with different scales of LUBM (Lehigh University Benchmark) RDF benchmark data sets. We make a comparison of Acacia-RDF's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported less than ten seconds elapsed times on 16 places for running the first and the third queries of the LUBM benchmark on LUBM 160 universities data set with 3.6 million vertices and 28.5 million edges which was 1.7GB in size. Through this work we describe and demonstrate the use of X10 language for development of scalable RDF graph data management systems.
-
introducing acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Parallel and Distributed Processing Symposium, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in HPC graph mining in recent years. However, the existing RDF Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we demonstrate the implementation of scalable graph algorithms such as Triangle counting with such partitioned data sets. We present performance results gathered from Acacia with different scales of LUBM RDF benchmark data sets and make a comparison of Acacia's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported approximately 2 seconds elapsed time on 4 places for running the first and third queries of the LUBM benchmark on LUBM scale 40 data set. Through this work we introduce the use of X10 language for scalable RDF graph data management.
-
towards scalable distributed graph Database Engine for hybrid clouds
Proceedings of the 5th International Workshop on Data-Intensive Computing in the Clouds, 2014Co-Authors: Miyuru Dayarathna, Toyotaro SuzumuraAbstract:Large graph data management and mining in clouds has become an important issue in recent times. We propose Acacia which is a distributed graph Database Engine for scalable handling of such large graph data. Acacia operates between the boundaries of private and public clouds. Acacia partitions and stores the graph data in the private cloud during its initial deployment. Acacia bursts into the public cloud when the resources of the private cloud are insufficient to maintain its service-level agreements. We implement Acacia using X10 programming language. We describe how Top-K PageRank has been implemented in Acacia. We report preliminary experiment results conducted with Acacia on a small compute cluster. Acacia is able to upload 69 million edges LiveJournal social network data set in about 10 minutes. Furthermore, Acacia calculates the average out degree of vertices of LiveJournal graph in 2 minutes. These results indicate Acacias potential for handling large graphs.
Sanath Jayasena - One of the best experts on this subject based on the ideXlab platform.
-
An X10-Based Distributed Streaming Graph Database Engine
2017 IEEE 24th International Conference on High Performance Computing (HiPC), 2017Co-Authors: Miyuru Dayarathna, Sanath Jayasena, Sathya Bandara, Nandula Jayamaha, Mahen Herath, Achala Madhushan, Toyotaro SuzumuraAbstract:Streaming graph data mining has become a significant issue in high performance graph mining due to the increasing appearance of graph data sets as streams. In this paper we propose Acacia-Stream which is a scalable distributed streaming graph Database Engine developed with X10 programming language. Graph streams are partitioned using a streaming graph partitioner algorithm in Acacia-Stream and streaming graph processing queries are run on the graph streams. The partitioned data sets are persisted on secondary storage across X10 places. We investigate on the use of three different streaming graph partitioner algorithms called hash, Linear Deterministic Greedy, and Fennel algorithms and report their performance. Furthermore, to demonstrate Acacia-Stream's streaming graph processing capabilities we implement streaming triangle counting with Acacia-Stream. We present performance results gathered from Acacia-Stream with different large scale streaming data sets in both horizontal and vertical scalability experiments. Furthermore, we compare streaming graph loading performance of Acacia-Stream with Neo4j and Oracle's PGX graph Database servers. From these experiments we observed that Acacia-Stream's Fennel partitioner based graph uploader can upload a 948MB rmat22 graph in 1283.42 seconds which is 38% faster than PGX graph Database server and 12.8 times faster than Neo4j Database server. Acacia-Stream's Streaming Partitioner's batch size adjustments based optimizations reduced the time used by the network communications almost by half.
-
acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Conference on Cloud Computing, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in High Performance graph mining in recent years. However, the existing Resource Description Framework (RDF) Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL (an RDF query language) processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we designed and implemented a replication based fault tolerance mechanism for Acacia-RDF. We present performance results gathered from Acacia with different scales of LUBM (Lehigh University Benchmark) RDF benchmark data sets. We make a comparison of Acacia-RDF's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported less than ten seconds elapsed times on 16 places for running the first and the third queries of the LUBM benchmark on LUBM 160 universities data set with 3.6 million vertices and 28.5 million edges which was 1.7GB in size. Through this work we describe and demonstrate the use of X10 language for development of scalable RDF graph data management systems.
-
introducing acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Parallel and Distributed Processing Symposium, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in HPC graph mining in recent years. However, the existing RDF Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we demonstrate the implementation of scalable graph algorithms such as Triangle counting with such partitioned data sets. We present performance results gathered from Acacia with different scales of LUBM RDF benchmark data sets and make a comparison of Acacia's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported approximately 2 seconds elapsed time on 4 places for running the first and third queries of the LUBM benchmark on LUBM scale 40 data set. Through this work we introduce the use of X10 language for scalable RDF graph data management.
Sameera Nandasiri - One of the best experts on this subject based on the ideXlab platform.
-
acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Conference on Cloud Computing, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in High Performance graph mining in recent years. However, the existing Resource Description Framework (RDF) Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL (an RDF query language) processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we designed and implemented a replication based fault tolerance mechanism for Acacia-RDF. We present performance results gathered from Acacia with different scales of LUBM (Lehigh University Benchmark) RDF benchmark data sets. We make a comparison of Acacia-RDF's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported less than ten seconds elapsed times on 16 places for running the first and the third queries of the LUBM benchmark on LUBM 160 universities data set with 3.6 million vertices and 28.5 million edges which was 1.7GB in size. Through this work we describe and demonstrate the use of X10 language for development of scalable RDF graph data management systems.
-
introducing acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Parallel and Distributed Processing Symposium, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in HPC graph mining in recent years. However, the existing RDF Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we demonstrate the implementation of scalable graph algorithms such as Triangle counting with such partitioned data sets. We present performance results gathered from Acacia with different scales of LUBM RDF benchmark data sets and make a comparison of Acacia's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported approximately 2 seconds elapsed time on 4 places for running the first and third queries of the LUBM benchmark on LUBM scale 40 data set. Through this work we introduce the use of X10 language for scalable RDF graph data management.
Isuru Herath - One of the best experts on this subject based on the ideXlab platform.
-
acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Conference on Cloud Computing, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in High Performance graph mining in recent years. However, the existing Resource Description Framework (RDF) Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL (an RDF query language) processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we designed and implemented a replication based fault tolerance mechanism for Acacia-RDF. We present performance results gathered from Acacia with different scales of LUBM (Lehigh University Benchmark) RDF benchmark data sets. We make a comparison of Acacia-RDF's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported less than ten seconds elapsed times on 16 places for running the first and the third queries of the LUBM benchmark on LUBM 160 universities data set with 3.6 million vertices and 28.5 million edges which was 1.7GB in size. Through this work we describe and demonstrate the use of X10 language for development of scalable RDF graph data management systems.
-
introducing acacia rdf an x10 based scalable distributed rdf graph Database Engine
International Parallel and Distributed Processing Symposium, 2016Co-Authors: Miyuru Dayarathna, Isuru Herath, Yasima Dewmini, Gayan Mettananda, Sameera Nandasiri, Sanath Jayasena, Toyotaro SuzumuraAbstract:Linked data mining has become one of the key questions in HPC graph mining in recent years. However, the existing RDF Database Engines are not scalable and are less reliable in heterogeneous clouds. In this paper we describe the design and implementation of Acacia-RDF which is a scalable distributed RDF graph Database Engine developed with X10 programming language to solve this issue. Acacia-RDF partitions the RDF data sets into subgraphs following vertex cut paradigm. The partitioned data sets are persisted on secondary storage across X10 places. We developed a scalable SPARQL processor for Acacia-RDF which operates on top of partitioned RDF data. Furthermore, we demonstrate the implementation of scalable graph algorithms such as Triangle counting with such partitioned data sets. We present performance results gathered from Acacia with different scales of LUBM RDF benchmark data sets and make a comparison of Acacia's performance against Neo4j graph Database server. From the scalability experiments conducted upto 16 X10 places, we observed that Acacia-RDF scales well with LUBM data sets. Acacia-RDF reported approximately 2 seconds elapsed time on 4 places for running the first and third queries of the LUBM benchmark on LUBM scale 40 data set. Through this work we introduce the use of X10 language for scalable RDF graph data management.
