Fabien Delalondre

Reconstruction and Simulation of Neocortical Microcircuitry

UNLABELLED We present a first-draft digital reconstruction of the microcircuitry of somatosensory cortex of juvenile rat. The reconstruction uses cellular and synaptic organizing principles to algorithmically reconstruct detailed anatomy and physiology from sparse experimental data. An objective anatomical method defines a neocortical volume of 0.29 ± 0.01 mm(3) containing ~31,000 neurons, and patch-clamp studies identify 55 layer-specific morphological and 207 morpho-electrical neuron subtypes. When digitally reconstructed neurons are positioned in the volume and synapse formation is restricted to biological bouton densities and numbers of synapses per connection, their overlapping arbors form ~8 million connections with ~37 million synapses. Simulations reproduce an array of in vitro and in vivo experiments without parameter tuning. Additionally, we find a spectrum of network states with a sharp transition from synchronous to asynchronous activity, modulated by physiological mechanisms. The spectrum of network states, dynamically reconfigured around this transition, supports diverse information processing strategies. PAPERCLIP VIDEO ABSTRACT.

The neocortical microcircuit collaboration portal: a resource for rat somatosensory cortex.

We have established a multi-constraint, data-driven process to digitally reconstruct, and simulate prototypical neocortical microcircuitry, using sparse experimental data. We applied this process to reconstruct the microcircuitry of the somatosensory cortex in juvenile rat at the cellular and synaptic levels. The resulting reconstruction is broadly consistent with current knowledge about the neocortical microcircuit and provides an array of predictions on its structure and function. To engage the community in exploring, challenging, and refining the reconstruction, we have developed a collaborative, internet-accessible facility-the Neocortical Microcircuit Collaboration portal (NMC portal; https://bbp.epfl.ch/nmc-portal). The NMC portal allows users to access the experimental data used in the reconstruction process, download cellular and synaptic models, and analyze the predicted properties of the microcircuit: six layers, similar to 31,000 neurons, 55 morphological types, 11 electrical types, 207 morpho-electrical types, 1941 unique synaptic connection types between neurons of specific morphological types, predicted properties for the anatomy and physiology of similar to 40 million intrinsic synapses. It also provides data supporting comparison of the anatomy and physiology of the reconstructed microcircuit against results in the literature. The portal aims to catalyzee consensus on the cellular and synaptic organization of neocortical microcircuitry (ion channel, neuron and synapse types and distributions, connectivity, etc.). Community feedback will contribute to refined versions of the reconstruction to be released periodically. We consider that the reconstructions and the simulations they enable represent a major step in the development of in silica neuroscience.

Neighborhood communication paradigm to increase scalability in large-scale dynamic scientific applications

This paper introduces a general-purpose communication package built on top of MPI which is aimed at improving inter-processor communications independently of the supercomputer architecture being considered. The package is developed to support parallel applications that rely on computation characterized by large number of messages of various sizes, often small, that are focused within processor neighborhoods. In some cases, such as solvers having static mesh partitions, the number and size of messages are known a priori. However, in other cases such as mesh adaptation, the messages evolve and vary in number and size and include the dynamic movement of partition objects. The current package provides a utility for dynamic applications based on two key attributes that are: (i) explicit consideration of the neighborhood communication pattern to avoid many-to-many calls and also to reduce the number of collective calls to a minimum, and (ii) use of non-blocking MPI functions along with message packing to manage message flow control and reduce the number and time of communication calls. The test application demonstrated is parallel unstructured mesh adaptation. Results on IBM Blue Gene/P and Cray XE6 computers show that the use of neighborhood-based communication control leads to scalable results when executing generally imbalanced mesh adaptation runs.

Rebasing I/O for Scientific Computing: Leveraging Storage Class Memory in an IBM BlueGene/Q Supercomputer

Storage class memory is receiving increasing attention for use in HPC systems for the acceleration of intensive IO operations. We report a particular instance using SLC FLASH memory integrated with an IBM BlueGene/Q supercomputer at scale Blue Gene Active Storage, BGAS. We describe two principle modes of operation of the non-volatile memory: 1 block device; 2 direct storage access DSA. The block device layer, built on the DSA layer, provides compatibility with IO layers common to existing HPC IO systems POSIX, MPIO, HDF5 and is expected to provide high performance in bandwidth critical use cases. The novel DSA strategy enables a low-overhead, byte addressable, asynchronous, kernel by-pass access method for very high user space IOPs in multithreaded application environments. Here, we expose DSA through HDF5 using a custom file driver. Benchmark results for the different modes are presented and scale-out to full system size showcases the capabilities of this technology.

Key/Value-Enabled Flash Memory for Complex Scientific Workflows with On-Line Analysis and Visualization

Scientific workflows are often composed of compute-intensive simulations and data-intensive analysis and visualization, both equally important for productivity. High-performance computers run the compute-intensive phases efficiently, but data-intensive processing is still getting less attention. Dense non-volatile memory integrated into super-computers can help address this problem. In addition to density, it offers significantly finer-grained I/O than disk-based I/O systems. We present a way to exploit the fundamental capabilities of Storage-Class Memories (SCM), such as Flash, by using scalable key-value (KV) I/O methods instead of traditional file I/O calls commonly used in HPC systems. Our objective is to enable higher performance for on-line and near-line storage for analysis and visualization of very high resolution, but correspondingly transient, simulation results. In this paper, we describe 1) the adaptation of a scalable key-value store to a BlueGene/Q system with integrated Flash memory, 2) a novel key-value aggregation module which implements coalesced, function-shipped calls between the clients and the servers, and 3) the refactoring of a scientific workflow to use application-relevant keys for fine-grained data subsets. The resulting implementation is analogous to function-shipping of POSIX I/O calls but shows an order of magnitude increase in read and a factor 2.5x increase in write IOPS performance (11 million read IOPS, 2.5 million write IOPS from 4096 compute nodes) when compared to a classical file system on the same system. It represents an innovative approach for the integration of SCM within an HPC system at scale.

Performance evaluation of the IBM POWER8 architecture to support computational neuroscientific application using morphologically detailed neurons

A part of large-scale future supercomputing architectures will rely on a combination of IBM POWER9 and Nvidia GPU processors. In order to anticipate software design, implementation and optimization decisions and optimally prepare scientific applications targetting exascale computing, it is mandatory to already analyze application performance on prototype hardware. In this article an evaluation of the performance of the IBM POWER8 system is provided in the context of NEURON software, a widely-used computational neuroscientific application modeling large-scale neuronal network using detailed morphologies. This evaluation describes in details how to accurately measure the performance of the system but also provides a first detailed performance analysis of representative kernels of NEURON. From this analysis, improvements can be suggested at the level of the application but also at the level of the performance measurement and system setup.

Cyme: A Library Maximizing SIMD Computation on User-Defined Containers

This paper presents Cyme, a C++ library aiming at abstracting the usage of SIMD instructions while maximizing the usage of the underlying hardware. Unlike similar efforts such as Boost.simd or VC, Cyme provides generic high level containers to the users which hides SIMD complexity. Cyme accomplishes this by 1 optimization of the Abstract Syntax Tree using Expression Templates Programming to prevent temporary copies and maximize the use of Fuse Multiply Add instructions and 2 creating a data layout in memory AoS or AoSoA, which minimizes data addressing and manipulation throughout all SIMD registers. Implementation of Cyme library has been accomplished on the IBM Blue Gene/Q architecture using the 256 bit SIMD extensions QPX of the Power A2 processor. Functionality of the library is demonstrated on a computationally intensive kernel of a neuro-scientific application where an increase of GFlop/s performance by a factor of 6.72 over the original implementation is observed using Clang compiler.

Nix based fully automated workflows and ecosystem to guarantee scientific result reproducibility across software environments and systems

Reproducibility is a key requirement to scientific development. Any scientific process, including software simulations, must be able to be replicated in order to prove the robustness of its process and the validity of its results. If an approach based on the extensive documentation of the process itself maybe considered as sufficient to guarantee reproducibility of results in domains like Physics or Biology, such a requirement proves to be incomplete for software being executed on high performance computing platforms. The specifics of the customized and exotic HPC architectures, the fast evolution of the software development environment as well as the various variables that can pollute the software development and building process are just few of the many possible sources of scientific result corruption. We describe in this paper how the developers of the Blue Brain Project built a software development ecosystem based on the Nix packaging and build system in order to guarantee the full portability, traceability and reproducibility of scientific results.

Neuromapp: A Mini-application Framework to Improve Neural Simulators

The increasing complexity and heterogeneity of extreme scale systems makes the optimization of large scale scientific applications particularly challenging. Efficiently leveraging these complex systems requires a great deal of technical expertise and a considerable amount of man-hours. The computational neuroscience community relies on an handful of those frameworks to model the electrical activity of brain tissue at different scales. As the members of the Blue Brain Project actively contribute to a large part of those frameworks, it becomes mandatory to implement a strategy to reduce the overall development cost. Therefore, we present Neuromapp, a computational neuroscience mini-application framework. Neuromapp consists of a number of mini-apps (small standalone applications) that represent a single functionality in one of the large scientific frameworks. The collection of several mini-apps forms a skeleton which is able to reproduce the original workflow of the scientific application. Thus, it becomes easy to investigate both single component and workflow optimizations, new software and hardware systems or future system design. New solutions can then be integrated into the large scientific applications if proved to be successful, reducing the overall development and optimization effort.

Accelerating Data Analysis in Simulation Neuroscience with Big Data Technologies

Important progress in computational sciences has been made possible recently thanks to the increasing computing power of high performance systems. Following this trend, larger scientific studies, like brain tissue simulations, will continue to grow in the future. In addition to the challenges of conducting these experiments, we foresee an explosion of the amount of data generated and the consequent unfeasibility of analyzing and understanding the results with the current techniques.