Memcomputing is a novel computing paradigm that employs time non-local dynamical systems to compute with and in memory. The digital version of these machines [digital memcomputing machines or (DMMs)] is scalable, and is particularly suited to solve combinatorial optimization problems. One of its possible realizations is by means of standard electronic circuits, with and without memory. Since these elements are non-quantum, they can be described by ordinary differential equations. Therefore, the circuit representation of DMMs can also be simulated efficiently on our traditional computers. We have indeed previously shown that these simulations only require time and memory resources that scale linearly with the problem size when applied to finding a good approximation to the optimum of hard instances of the maximum-satisfiability problem. The state-of-the-art algorithms, instead, require exponential resources for the same instances. However, in that work, we did not push the simulations to the limit of the processor used. Since linear scalability at smaller problem sizes cannot guarantee linear scalability at much larger sizes, we have extended these results in a stress-test up to $64\times 10^{6}$ variables (corresponding to about 1 billion literals), namely the largest case that we could fit on a single core of an Intel Xeon E5-2860 with 128 GB of dynamic random-access memory (DRAM). For this test, we have employed a commercial simulator, Falcon of MemComputing, Inc. We find that the simulations of DMMs still scale linearly in both time and memory up to these very large problem sizes versus the exponential requirements of the state-of-the-art solvers. These results further reinforce the advantages of the physics-based memcomputing approach compared with traditional ones.

Data analytics has become an integral part of large-scale scientific computing. Among various data analytics frameworks, MapReduce has gained the most traction. Although some efforts have been made to enable efficient MapReduce for supercomputing systems, they are often limited to fairly homogeneous workloads where equal partitioning of input data across tasks results in essentially equal output or temporary data generated on each task. For workloads that are more skewed, however, current implementations can result in imbalance in memory usage and, consequently, can cause a slowdown in execution time and a loss in data scalability. To tackle this problem, we enhance a previously published memory-conscious MapReduce over MPI framework called Mimir. Our enhancements to Mimir include combiner and dynamic repartition optimizations to minimize and balance memory usage and to achieve close to optimal balance of the memory usage across processes and to reduce the execution time by up to 12 times. Experimental results show that Mimir can scale to at least 3072 processes on the Tianhe-2 supercomputer on skewed datasets.

Analyzing large volumes of data is becoming more and more important in various scientific computing domains. MapReduce over MPI frameworks are an appealing solution to enable scalable big data analytics on supercomputing systems. These systems can further leverage features of MapReduce applications by merging (key/value) pairs before the reduce function in combiner optimizations. In this paper, we propose a pipeline combiner workflow and integrate it into Mimir, a cutting-edge implementation of Map Reduce over MPI. Our results with real datasets on the Tianhe-2 supercomputer prove that our pipeline combiner workflow can reduce memory usage up to 51% and improve the overall performance up to 61%.

K-mer counting is a fundamental operation in DNA research and genome analytics; its application includes estimating genome assembly, understanding similarities in genomic samples, and merging a newly processed genome with a reference genome. As the genome dataset becomes larger and larger, designing a highly optimized distributed-memory implementation becomes more and more important. Current distributed-memory solutions have two limitations: they have a high memory footprint, and they do not provide advanced optimizations for loading enormous genome datasets into memory. Based on these observations, we present Bloomfish, a distributed, memory-efficient, scalable solution to the limits of current work. To keep a low memory footprint, Bloomfish leverages the compact hash array design of the single-node Jellyfish system and the optimized workflow of the high-performance MapReduce framework Mimir. We have also codesigned Mimir's I/O to efficiently load enormous datasets. We ran Bloomfish on the Tianhe-2 supercomputer with large sequence datasets (up to 24 TB). Our results show that Bloomfish achieves unprecedented scalability in genome analytics.

In this paper we present Mimir, a new implementation of MapReduce over MPI. Mimir inherits the core principles of existing MapReduce frameworks, such as MR-MPI, while redesigning the execution model to incorporate a number of sophisticated optimization techniques that achieve similar or better performance with significant reduction in the amount of memory used. Consequently, Mimir allows significantly larger problems to be executed in memory, achieving large performance gains. We evaluate Mimir with three benchmarks on two highend platforms to demonstrate its superiority compared with that of other frameworks.

Traditional scientific and emerging data analytics applications require fast, power-efficient, large, and persistent memories. Combining all these characteristics within a single memory technology is expensive and hence future supercomputers will feature different memory technologies side-by-side. However, it is a complex task to program hybrid-memory systems and to identify the best object-to-memory mapping. We envision that programmers will probably resort to use default configurations that only require minimal interventions on the application code or system settings. In this work, we argue that intelligent, fine-grained data placement can achieve higher performance than default setups.

We present a scalable method to extensively search for and accurately select pharmaceutical drug candidates in large spaces of drug conformations computationally generated and stored across the nodes of a large distributed system. For each legend conformation in the dataset, our method first extracts relevant geometrical properties and transforms the properties into a single metadata point in the three-dimensional space. Then, it performs an ochre-based clustering on the metadata to search for predominant clusters. Our method avoids the need to move legend conformations among nodes because it extracts relevant data properties locally and concurrently. By doing so, we can perform accurate and scalable distributed clustering analysis on large distributed datasets. We scale the analysis of our pharmaceutical datasets a factor of 400X higher in performance and 500X larger in size than ever before. We also show that our clustering achieves higher accuracy compared with that of traditional clustering methods and conformational scoring based on minimum energy.

Meeting the 20MW power envelope sought for exascale is one of the greatest challenges in designing those class of systems. Addressing this challenge requires over-provisioned and dynamically reconfigurable system with fine-grained control on power and speed of the individual cores. In this paper, we present EfficientSpeed (ES), a library that improves energy efficiency in scientific computing by carefully selecting the speed of the processor. The run-time component of ES adjusts the speed of the processor (via DVFS and clock modulation) dynamically while preserving the desired level of the performance. These adjustments are based on online performance and energy measurements, user-selected policies that dictate the aggressiveness of adjustments, and user-defined performance requirements. Our results quantify the best energy savings that can be achieved by controlling the speed of the processor, with today's technology, at the cost of negligible performance degradation. We then demonstrate that ES is effective in automatically calibrating the speed of execution in real applications, saving energy and meeting the desired performance goal.We evaluate ES on GAMESS, an ab initio quantum chemistry package. We show that ES respects the stipulated 5% performance loss bound and achieves 16% decrease in energy required to complete the execution while running with a power draw that is 18% lower.

DRAM technology has several shortcomings in terms of performance, energy efficiency and scaling. Several emerging memory technologies have the potential to compensate for the limitations of DRAM when replacing or complementing DRAM in the memory sub-system. In this paper, we evaluate the impact of emerging technologies on HPC and data-intensive workloads modeling a 5-level hybrid memory hierarchy design. Our results show that 1) an additional level of faster DRAM technology (i.e. EDRAM or HMC) interposed between the last level cache and DRAM can improve performance and energy efficiency, 2) a non-volatile main memory (i.e. PCM, STTRAM, or FeRAM) with a small DRAM acting as a cache can reduce the cost and energy consumption at large capacities, and 3) a combination of the two approaches, which essentially replaces the traditional DRAM with a small EDRAM or HMC cache between the last level cache and the non-volatile memory, can grant capacity and improved performance and energy efficiency. We also explore a hybrid DRAM-NVM design with a partitioned address space and find that this approach is marginally beneficial compared to the simpler 5-level design. Finally, we generalize our analysis and show the impact of emerging technologies for a range of latency and energy parameters.

This paper presents a one-pass, distributed method that enables in-situ data analysis for large protein folding trajectory datasets by executing sufficiently fast, avoiding moving trajectory data, and limiting the memory usage. First, the method extracts the geometric shape features of each protein conformation in parallel. Then, it classifies sets of consecutive conformations into meta-stable and transition stages using a probabilistic hierarchical clustering method. Lastly, it rebuilds the global knowledge necessary for the intraand inter-trajectory analysis through a reduction operation. The comparison of our method with a traditional approach for a villin headpiece sub domain shows that our method generates significant improvements in execution time, memory usage, and data movement. Specifically, to analyze the same trajectory consisting of 20,000 protein conformations, our method runs in 41.5 seconds while the traditional approach takes approximately 3 hours, uses 6.9MB memory per core while the traditional method uses 16GB on one single node where the analysis is performed, and communicates only 4.4KB while the traditional method moves the entire dataset of 539MB. The overall results in this paper support our claim that our method is suitable for in-situ data analysis of folding trajectories.

We present Bamboo, a custom source-to-source translator that transforms MPI C source into a data-driven form that automatically overlaps communication with available computation. Running on up to 98304 processors of NERSC's Hopper system, we observe that Bamboo's overlap capability speeds up MPI implementations of a 3D Jacobi iterative solver and Cannon's matrix multiplication. Bamboo's generated code meets or exceeds the performance of hand optimized MPI, which includes split-phase coding, the method classically employed to hide communication. We achieved our results with only modest amounts of programmer annotation and no intrusive reprogramming of the original application source.

The Gordon data intensive computing system was designed to handle problems with large memory requirements that cannot easily be solved using standard workstations or distributed memory supercomputers. We describe the unique features of Gordon that make it ideally suited for data mining and knowledge discovery applications: memory aggregation using the vSMP software solution from ScaleMP, I/O nodes containing 4 TB of low-latency flash memory, and a high performance parallel file system with 4 PB capacity. We also demonstrate how a number of standard data mining tools (e.g. Matlab, WEKA, R) can be used effectively on Dash, an early prototype of the full Gordon system.

Tarragon is an actor-based programming model and library for implementing latency tolerant asynchronous event driven simulations. It is novel in its support for meta data describing run time virtualized process structures, which may be optimized as a free-standing object. We demonstrate early results with a synthetic benchmark, and observe that Tarragon can mask communication costs with ongoing computation

Tarragon is an actor-based programming model and library for implementing parallel scientific applications requiring fine grain asynchronous communication. Tarragon raises the level of abstraction by encapsulating run-time services that mange the actor semantics. The workload is over-decomposed into many virtual processes called WorkUnits. WorkUnits can become ready for execution after receiving input; scheduling and communication services coordinate WorkUnit execution and management. In order to maintain balanced workloads, Tarragon automatically monitors workload distribution and redistributes as needed. Tarragon is novel in its support for meta data describing run-time virtual process structures used to manage actor semantics. This meta data may be used to guide run time services policies in order to optimize performance. We are currently applying Tarragon to the MCell cell microphysiology simulator and are considering other applications as well, such as sparse matrix linear algebra.

Summary form only given. Accurate, continuous resource monitoring and profiling are critical for enabling performance tuning and scheduling optimization. In desktop grid systems that employ sandboxing, these issues are challenging because (1) subjobs inside sandboxes are executed in a virtual computing environment and (2) the state of the virtual computing environment within the sandboxes is reset to empty after each subjob completes. DGMonitor is a monitoring tool, which builds a global, accurate, and continuous view of real resource utilization for desktop grids with sandboxing. Our monitoring tool measures performance unobtrusively and reliably, uses a simple performance data model, and is easy to use. Our measurements demonstrate that DGMonitor can scale to large desktop grids (up to 12000 workers) with low monitoring overhead in terms of resource consumption (less than 0.1%) on desktop PCs. Though we developed DGMonitor with the Entropia DCGrid platform, our tool is easily integrated into other desktop grid systems. In all of these systems, DGMonitor data can support existing and novel information services, particularly for performance tuning and scheduling.