Jiazheng Zhou

Data adapter for querying and transformation between SQL and NoSQL database

As the growing of applications with big data in cloud computing become popular, many existing systems expect to expand their service to support the explosive increase of data. We propose a data adapter system to support hybrid database architecture including a relational database (RDB) and NoSQL database. It can support query from application and deal with database transformation at the same time. We provide three modes of query approach in data adapter system: blocking transformation mode (BT mode), blocking dump mode (BD mode), and direct access mode (DA mode). We provide a data synchronization mechanism and describe the design and implementation in detail. This paper focuses on velocity with proposed three modes and partly variety with data stored in RDB, NoSQL database and temporary files. With the proposed data adapter system, we can provide a seamless mechanism to use RDB and NoSQL database at the same time. This paper presents data adapter to make possible the automated transformation of multi-structured data in Relational Database (RDB) and NoSQL systems.With the proposed data adapter, a seamless mechanism is provided for constructing hybrid database systems.With the proposed data adapter, hybrid database systems can be performed in an elastic manner, i.e., access can be either RDB or NoSQL, depending on the size of data.

HIERARCHICAL MAPPING FOR HPC APPLICATIONS

As the high performance computing systems scale up, mapping the tasks of a parallel application onto physical processors to allow efficient communication becomes one of the critical performance issues. Existing algorithms were usually designed to map applications with regular communication patterns. Their mapping criterion usually overlooks the size of communicated messages, which is the primary factor of communication time. In addition, most of their time complexities are too high to process large scale problems. In this paper, we present a hierarchical mapping algorithm (HMA), which is capable of mapping applications with irregular communication patterns. It first partitions tasks according to their run-time communication information. The tasks that communicate with each others more frequently are regarded as strongly connected. Based on their connectivity strength, the tasks are partitioned into super nodes based on the algorithms in spectral graph theory. The hierarchical partitioning reduces the mapping algorithm complexity to achieve scalability. Finally, the run-time communication information will be used again in fine tuning to explore better mappings. With the experiments, we show how the mapping algorithm helps to reduce the point-to-point communication time for the PDGEMM, a ScaLAPACK matrix multiplication computation kernel, up to 20% and the AMG2006, a tier 1 application of the Sequoia benchmark, up to 7%.

Hierarchical Mapping for HPC Applications

As the high performance computing systems scale up, mapping the tasks of a parallel application onto physical processors to allow efficient communication becomes one of the critical performance issues. Existing algorithms were usually designed to map applications with regular communication patterns. Their mapping criterion usually overlooks the size of communicated messages, which is the primary factor of communication time. In addition, most of their time complexities are too high to process large scale problems. In this paper, we present a hierarchical mapping algorithm (HMA), which is capable of mapping applications with irregular communication patterns. It first partitions tasks according to their run-time communication information. The tasks that communicate with each others more frequently are regarded as strongly connected. Based on their connectivity strength, the tasks are partitioned into super nodes based on the algorithms in spectral graph theory. The hierarchical partitioning reduces the mapping algorithm complexity to achieve scalability. Finally, the run-time communication information will be used again in fine tuning to explore better mappings. With the experiments, we show how the mapping algorithm helps to reduce the point-to-point communication time for the PDGEMM, a ScaLAPACK matrix multiplication computation kernel, up to 20% and the AMG2006, a tier 1 application of the Sequoia benchmark, up to 7%.

Hardware supported multicast in fat-tree-based InfiniBand networks

Abstract The multicast operation is a very commonly used operation in parallel applications. It can be used to implement many collective communication operations as well. Therefore, its performance will affect parallel applications and collective communication operations. With the hardware supported multicast of the InfiniBand Architecture (IBA), in this paper, we propose a cyclic multicast scheme for fat-tree-based (m-port n-tree) InfiniBand networks. The basic concept of the proposed cyclic multicast scheme is to find the union sets of the output ports of switches in the paths between the source processing node and each destination processing node in a multicast group. Based on the union sets and the path selection scheme, the forwarding table for a given multicast group can be constructed. We implement the proposed multicast scheme along with the OpenSM multicast scheme and the unicast scheme on an m-port n-tree InfiniBand network simulator. Several one-to-many, many-to-many, many-to-all, and all-to-many multicast cases are simulated. The simulation results show that the proposed multicast scheme outperforms the unicast scheme for all simulated cases. For one-to-many case, the performance of the cyclic multicast scheme is the same as that of the OpenSM multicast scheme. For many-to-many and all-to-many cases, the cyclic multicast scheme outperforms the OpenSM multicast scheme. For many-to-all case, the performance of the cyclic multicast scheme is a little better than that of the OpenSM multicast scheme.

Tree-turn routing: an efficient deadlock-free routing algorithm for irregular networks

In this paper, we propose a general turn model, called a Tree-turn model, for tree-based routing algorithms on irregular topologies. In the Tree-turn model, links are classified as either a tree link or a cross link and six directions are associated with the channels of links. Then we can prohibit some of the turns formed by these six directions such that an efficient deadlock-free routing algorithm, Tree-turn routing, can be derived. There are three phases to develop the Tree-turn routing. First, a coordinated tree for a given topology is created. Second, a communication graph is constructed based on the topology and the corresponding coordinated tree. Third, the forwarding table is set up by using all-pairs shortest path algorithm according to the prohibited turns in the Tree-turn model and the directions of the channels in the communication graph. To evaluate the performance of the proposed Tree-turn routing, we develop a simulator and implement Tree-turn routing along with up*/down* routing, L-turn routing, and up*/down* routing with DFS methodology. The simulation results show that Tree-turn routing outperforms other routing algorithms for all the test cases.

A Tree-Turn Model for Irregular Networks

In this paper, we propose a general turn model, Tree-turn model, for irregular topology. In Tree-turn model, links are classified as either tree or cross and six directions are associated with channels of links. From these six directions, we prohibit some turns such that an efficient deadlock-free routing algorithm, Tree-turn routing, can be derived. There are three phases to construct the Tree-turn routing. First, build up a coordinated tree for a given topology. Second, construct a communication graph of the topology and the corresponding coordinated tree. Third, set up the forwarding table by using the all-pairs shortest path algorithm according to the prohibited turns derived from the Tree-turn model and the directions of the channels in communication graph. To evaluate the performance, we implement the Tree-turn routing algorithm along with the up*/down* routing algorithm and the L-turn routing algorithm on a software simulator. The simulation results show that Tree-turn routing outperforms other two routing algorithms for all test cases

Multicast in Fat-Tree-Based InfiniBand Networks

The multicast operation is a very commonly used operation in parallel applications. With the hardware supported multicast of the InfiniBand architecture (IBA), we propose a cyclic multicast scheme for fat-tree-based (m-port n-tree) InfiniBand networks. The basic concept of the proposed cyclic multicast scheme is to find the union sets of the output ports of switches in the paths between the source processing node and each destination processing node in a multicast group. Based on the union sets and the path selection scheme, the forwarding table for a given multicast group can be constructed. We implement the proposed multicast scheme along with the OpenSM multicast scheme and the unicast scheme on an m-port n-tree InfiniBand network simulator. The simulation results show that the proposed multicast scheme outperforms the unicast scheme for all simulated cases. For many-to-many and all-to-many cases, the cyclic multicast scheme outperforms the OpenSM multicast scheme. For many-to-all case, the performance of the cyclic multicast scheme is a little better than that of the OpenSM multicast scheme

A hardware supported multicast scheme based on XY routing for 2-D mesh InfiniBand networks

The multicast operation is a useful operation in parallel applications. It is therefore important to ensure that for a given architecture, the parallel application runs efficiently. With the hardware-supported multicast of the InfiniBand Architecture (IBA), we propose a multicast scheme for m×n mesh InfiniBand networks based on XY routing. The basic concept of the proposed multicast scheme is to find the union sets of the output ports of switches, which are in the paths between the source node and each destination node in a multicast group. Furthermore, in the proposed scheme, we consider the usage of virtual lanes and evaluate their performance. We implement the proposed multicast scheme on a 2-D mesh InfiniBand network simulator. Several multicast configurations consisting of different message size, different traffic workload and different number of virtual lanes are simulated. The simulation results show that the proposed multicast scheme outperforms its corresponding unicast scheme for all simulation cases. The larger the message size, the larger the number of multicast source nodes, and the larger the size of the multicast group, the better the speedup that can be expected from the proposed multicast scheme. The usage of virtual lanes is also shown to improve the speed of the multicast operations.

Scalable Communication-Aware Task Mapping Algorithms for Interconnected Multicore Systems

Communication-aware task mapping algorithms, which map parallel tasks onto processing nodes according to the communication patterns of applications, are essential to reduce the communication time in modern high performance computing. In this paper, we design algorithms specifically for interconnected multicore systems, whose architectural property, namely small number of cores per node, large number of nodes, and large performance gap between the communication within a multicore and among multicores, had brought new challenges and opportunities to the mapping problem. Let k be the number of cores per multicore and n be the number of tasks. We consider the practical case that k is much smaller than n, for k = 2, 4, and 6. The designed algorithms are optimal for the mapping measurement, called Maximum Interconnective Message Size (MIMS), and of time complexity merely O(mlogm) for m communication pairs. Thus, they are highly scalable for large applications. We had experimented the algorithms on the IBM Blue Gene/P system for two synthetic benchmarks and two applications. The results show good communication performance improvement.

Hardware Supported Multicast in 2-D Mesh InfiniBand Networks

The multicast operation is a useful operation in parallel applications. With the hardware supported multicast of the InfiniBand Architecture (IBA), we propose a multicast scheme for mxn mesh InfiniBand networks based on the XY routing scheme. The basic concept of the proposed multicast scheme is to find the union sets of the output ports of switches that are in the paths between the source node and each destination node in a multicast group. We have implemented the proposed multicast scheme on a 2-D mesh InfiniBand network simulator. Several multicast cases with different message size and different traffic workload are simulated. The simulation results show that the proposed multicast scheme outperforms their corresponding unicast scheme for all simulated cases. The larger the message size, the number of multicast source nodes, and the size of the multicast group, the better speedup can be expected from the proposed multicast scheme.