Hierarchical task mapping of cell-based AMR cosmology simulations

Wu, Jingjin; Lan, Zhiling; Xiong, Xuanxing; Gnedin, Nickolay Y.; Kravtsov, Andrey V.

doi:10.1109/sc.2012.63

Cited by 12 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The space-filling curve domain decomposition approach we proposed in WS93 has been widely adopted in both application codes (e.g. [23,50,27,75]) and more general libraries [38,32]. Our claim that such orderings are also beneficial for improving memory hierarchy performance has also been validated [50,34].…”

Section: Domain Decompositionmentioning

confidence: 82%

2hot

Warren

2013

Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

We report on improvements made over the past two decades to our adaptive treecode N-body method (HOT). A mathematical and computational approach to the cosmological N-body problem is described, with performance and scalability measured up to 256k (2 18 ) processors. We present error analysis and scientific application results from a series of more than ten 69 billion (4096 3 ) particle cosmological simulations, accounting for 4 × 10 20 floating point operations. These results include the first simulations using the new constraints on the standard model of cosmology from the Planck satellite. Our simulations set a new standard for accuracy and scientific throughput, while meeting or exceeding the computational efficiency of the latest generation of hybrid TreePM N-body methods.

show abstract

Section: Domain Decompositionmentioning

confidence: 82%

2hot

Warren

2013

Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

show abstract

“…13,23 In this algorithm, the weighted communication graph G g is recursively split using a minimum edge-cut into equal halves, while the topology graph G t is split with a maximum edge-cut. 13,23 In this algorithm, the weighted communication graph G g is recursively split using a minimum edge-cut into equal halves, while the topology graph G t is split with a maximum edge-cut.…”

Section: Topology-aware Mappingmentioning

confidence: 99%

“…• Hop-bytes 11,23 measures the total size of data communication in bytes weighted by network distance. Network distance is measured as the length (number of network hops) of the shortest path between communicating processes in the network topology graph.…”

Section: Formalizing the Workflow Communication And Network Topologymentioning

confidence: 99%

“…The topology-aware mapping step uses a recursive bisection based algorithm. 13,23 In this algorithm, the weighted communication graph G g is recursively split using a minimum edge-cut into equal halves, while the topology graph G t is split with a maximum edge-cut. The runtime complexity of this algorithm is O(|E g |log(|V g |) + |E t |log(|V t |)).…”

Section: Grouping the Graph Partitionsmentioning

confidence: 99%

See 1 more Smart Citation

In‐memory staging and data‐centric task placement for coupled scientific simulation workflows

Zhang

Jin

Sun

et al. 2017

Concurrency and Computation

View full text Add to dashboard Cite

Coupled scientific simulation workflows are composed of heterogeneous component applications that simulate different aspects of the physical phenomena being modeled and that interact and exchange significant volumes of data at runtime. As the data volumes and generation rates keep growing, the traditional disk I/O-based data movement approach becomes cost prohibitive, and workflow requires more scalable and efficient approach to support the data movement. Moreover, the cost of moving large volume of data over system interconnection network becomes dominating and significantly impacts the workflow execution time. Minimize the amount of network data movement and localize data transfers are critical for reducing such cost. To achieve this, workflow task placement should exploit data locality to the extent possible and move computation closer to data. In this paper, we investigate applying in-memory data staging and data-centric task placement to reduce the data movement cost in large-scale coupled simulation workflows. Specifically, we present a distributed data sharing and task execution framework that (1) co-locates in-memory data staging on application compute nodes to store data that needs to be shared or exchanged and (2) uses data-centric task placement to map computations onto processor cores that a large portion of the data exchanges can be performed using the intra-node shared memory. We also present the implementation of the framework and its experimental evaluation on Titan Cray XK7 petascale supercomputer. KEYWORDScoupled simulations, data-intensive application workflows, data-centric task mapping, data staging 1 INTRODUCTION Emerging coupled simulation workflows are composed of multiple applications that interact and exchange data at runtime and have the potential to achieve higher accuracy and accelerate the data to insight process. Multi-physics multi-model simulation workflow simulates different aspects of the phenomena being modeled by coupling multiple physical models. For example, in the Community Earth System Model 1 application workflow, separate simulations are coupled to model the interaction of the earth's ocean, atmosphere, land surface, and sea ice. Meanwhile, online data analytics workflow supports analyzing raw simulation data while it is being generated. For example, online analytics workflow for combustion simulation S3D 2 extracts and streams simulation data to a number of analysis operations, eg, visualization, descriptive statistics, which execute concurrently and in parallel with the simulation.However, running coupled simulation workflows on extreme-scale computing system is non-trivial. One challenge of running coupled simulation workflow at scale is providing efficient data movement mechanism to extract data from 1 simulation, transfer, and redistribute the data to another simulation or analysis. Traditionally, workflow applications share and exchange data using distributed file systems. 3-6 But the increasing performance gap between computation and disk I/O introduces significan...

show abstract

“…Bhatelé et al introduce a framework based on heuristic models to map an application's communication graph to either 2D or 3D mesh-based topologies [16]. Wu et al propose a hierarchical task mapping for cosmological simulations, taking both the inter-node and intra-node mappings into account 110 [17]. Chung et al also propose a hierarchical mapping strategy that achieves O(n log n) complexity by grouping heavy-communicating processes together [18].…”

Section: Task Mapping For Communication-intensive Applicationsmentioning

confidence: 99%

Communication and cooling aware job allocation in data centers for communication-intensive workloads

Meng

Llamosí

Kaplan

et al. 2016

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Energy consumption is an increasingly important concern in data centers. Today, nearly half of the energy in data centers is consumed by the cooling infrastructure. Existing policies on thermally-aware workload allocation do not consider applications that include many tasks (or threads) running on a large set of nodes with significant communication among the tasks. Such jobs, however, constitute most of the cycles in high performance computing (HPC) domain, and have started to appear in other data centers as well. Job allocation strongly affects the performance of such communication-intensive applications. Communication-aware job allocation methods exist, but they focus solely on performance and do not consider cooling energy. This paper proposes a novel job allocation methodology to jointly minimize communication cost and cooling energy consumption in data centers. We formulate and solve the joint optimization problem using binary quadratic programming. Our joint optimization algorithm reduces cooling energy by 16.4% on average with only a 2.66% average increase in application running time compared to solely performance-aware allocations. To further optimize the communication cost, we develop a Charm++ based framework that extracts the communication behavior of applications. We then integrate our job allocation policy with recursive coordinate bisection (RCB) based task mapping method to place highly-communicating tasks in close proximity. Experimental results show that task mapping further decreases the communication cost by up to 20.9% compared to assuming all-to-all communication, a popular assumption in much of the prior work. 20 temperature is sufficient to cool the data center [4,5].Performance is another first-order constraint in data center design and management. Communication-intensive parallelized applications constitute most of HPC workloads, and have started to occupy more cycles in other data centers following the efforts in designing parallelized large-scale applications in many domains. These applications run 25 on multiple nodes with intensive communication between the threads, with job running times in the range of minutes, hours, or even days. In particular, the performance of a communication-intensive application is sensitive to the specific allocation and mapping onto the compute nodes. In other words, communication cost has a significant impact on system performance [6,7,8], with long communication distances and communication 30 resource contention increasing the overall delay. Leung et al. have shown a 2X slow down for two communication-intensive jobs when hand-placing them such that their communication paths overlap significantly [6].Most of the existing performance-aware job allocation algorithms focus on minimizing the average number of communication hops between communicating nodes (e.g., [7,8,35 9, 10]). For data centers running transactional enterprise loads, performance-aware job allocation algorithms focus on satisfying the response time constraints imposed by service level agreemen...

show abstract

Hierarchical task mapping of cell-based AMR cosmology simulations

Cited by 12 publications

References 23 publications

2hot

2hot

In‐memory staging and data‐centric task placement for coupled scientific simulation workflows

Communication and cooling aware job allocation in data centers for communication-intensive workloads

Contact Info

Product

Resources

About