Farhan Tauheed

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.

TOUCH: in-memory spatial join by hierarchical data-oriented partitioning

Efficient spatial joins are pivotal for many applications and particularly important for geographical information systems or for the simulation sciences where scientists work with spatial models. Past research has primarily focused on disk-based spatial joins; efficient in-memory approaches, however, are important for two reasons: a) main memory has grown so large that many datasets fit in it and b) the in-memory join is a very time-consuming part of all disk-based spatial joins. In this paper we develop TOUCH, a novel in-memory spatial join algorithm that uses hierarchical data-oriented space partitioning, thereby keeping both its memory footprint and the number of comparisons low. Our results show that TOUCH outperforms known in-memory spatial-join algorithms as well as in-memory implementations of disk-based join approaches. In particular, it has a one order of magnitude advantage over the memory-demanding state of the art in terms of number of comparisons (i.e., pairwise object comparisons), as well as execution time, while it is two orders of magnitude faster when compared to approaches with a similar memory footprint. Furthermore, TOUCH is more scalable than competing approaches as data density grows.

SCOUT: prefetching for latent structure following queries

Todays scientists are quickly moving from in vitro to in silico experimentation: they no longer analyze natural phenomena in a petri dish, but instead they build models and simulate them. Managing and analyzing the massive amounts of data involved in simulations is a major task. Yet, they lack the tools to efficiently work with data of this size. One problem many scientists share is the analysis of the massive spatial models they build. For several types of analysis they need to interactively follow the structures in the spatial model, e.g., the arterial tree, neuron fibers, etc., and issue range queries along the way. Each query takes long to execute, and the total time for executing a sequence of queries significantly delays data analysis. Prefetching the spatial data reduces the response time considerably, but known approaches do not prefetch with high accuracy. We develop SCOUT, a structure-aware method for prefetching data along interactive spatial query sequences. SCOUT uses an approximate graph model of the structures involved in past queries and attempts to identify what particular structure the user follows. Our experiments with neuro-science data show that SCOUT prefetches with an accuracy from 71% to 92%, which translates to a speedup of 4x-15x. SCOUT also improves the prefetching accuracy on datasets from other scientific domains, such as medicine and biology.

THERMAL-JOIN: A Scalable Spatial Join for Dynamic Workloads

Simulations have become ubiquitous in many domains of science. Today scientists study natural phenomena by first building massive three-dimensional spatial models and then by simulating the models at discrete intervals of time to mimic the behavior of natural phenomena. One frequently occurring challenge during simulations is the repeated computation of spatial self-joins of the model at each simulation time step. The join is performed to access a group of neighboring spatial objects (groups of particles, molecules or cosmological objects) so that scientists can calculate the cumulative effect (like gravitational force) on an object. Computing a self-join even in memory, soon becomes a performance bottleneck in simulation applications. The problem becomes even worse as scientists continue to improve the precision of simulations by increasing the number as well as the size (3D extent) of the objects. This leads to an exponential increase in join selectivity that challenges the performance and scalability of state-of-the-art approaches. We propose THERMAL-JOIN, a novel spatial self-join algorithm for dynamic memory-resident workloads. The algorithm groups objects in spatial proximity together into hot spots. Hot spots minimize the cost of computing join as objects assigned to a hot spot are guaranteed to overlap with each other. Using a nested spatial grid, THERMAL-JOIN partitions and indexes the dataset to locate hot spots. With experiments we show that our approach provides a speedup between 8 to 12x compared to the state of the art and also scales as scientists improve the precision of their simulations.

Data-driven neuroscience: enabling breakthroughs via innovative data management

Scientists in all disciplines increasingly rely on simulations to develop a better understanding of the subject they are studying. For example the neuroscientists we collaborate with in the Blue Brain project have started to simulate the brain on a supercomputer. The level of detail of their models is unprecedented as they model details on the subcellular level (e.g., the neurotransmitter). This level of detail, however, also leads to a true data deluge and the neuroscientists have only few tools to efficiently analyze the data. This demonstration showcases three innovative spatial management techniques that have substantial impact on computational neuroscience and other disciplines in that they allow to build, analyze and simulate bigger and more detailed models. More particularly, we demonstrate a tool that integrates three spatial data management techniques that have enabled breakthroughs in neuroscience: FLAT that enables efficient querying of spatial data, SCOUT that allows for fast exploration of spatial data and HiDOP that makes efficient data discovery possible.

BLOCK: Efficient Execution of Spatial Range Queries in Main-Memory

The execution of spatial range queries is at the core of many applications, particularly in the simulation sciences but also in many other domains. Although main memory in desktop and supercomputers alike has grown considerably in recent years, most spatial indexes supporting the efficient execution of range queries are still only optimized for disk access (minimizing disk page reads). Recent research has primarily focused on the optimization of known disk-based approaches for memory (through cache alignment etc.) but has not fundamentally revisited index structures for memory. In this paper we develop BLOCK, a novel approach to execute range queries on spatial data featuring volumetric objects in main memory. Our approach is built on the key insight that in-memory approaches need to be optimized to reduce the number of intersection tests (between objects and query but also in the index structure). Our experimental results show that BLOCK outperforms known in-memory indexes as well as in-memory implementations of disk-based spatial indexes up to a factor of 7. The experiments show that it is more scalable than competing approaches as the data sets become denser.

GIPSY: joining spatial datasets with contrasting density

Many scientific and geographical applications rely on the efficient execution of spatial joins. Past research has produced several efficient spatial join approaches and while each of them can join two datasets, the problem of efficiently joining two datasets with contrasting density, i.e., with the same spatial extent but with a wildly different number of spatial elements, has so far been overlooked. State-of-the-art data-oriented spatial join approaches (e.g., based on the R-Tree) suffer from degraded performance due to overlap, whereas space-oriented approaches excessively read data from disk. In this paper we develop GIPSY, a novel approach for the spatial join of two datasets with contrasting density. GIPSY uses fine-grained data-oriented partitioning and thus only retrieves the data needed for the join. At the same time it avoids the overlap related problems associated with data-oriented partitioning by using a crawling approach, i.e., without using a hierarchical tree. Our experiments show that GIPSY outperforms state-of-the-art disk-based spatial join algorithms by a factor of 2 to 18 and is particularly efficient when joining a dense dataset with several sparse datasets.

Configuring Spatial Grids for Efficient Main Memory Joins

The performance of spatial joins is becoming increasingly important in many applications, particularly in the scientific domain. Several approaches have been proposed for joining spatial datasets on disk and few in main memory. Recent results show that in main memory, grids are more efficient than the traditional tree based methods primarily developed for disk. The question how to configure the grid, however, has so far not been discussed.

OCTOPUS: Efficient query execution on dynamic mesh datasets

Scientists in many disciplines use spatial mesh models to study physical phenomena. Simulating natural phenomena by changing meshes over time helps to better understand the phenomena. The higher the precision of the mesh models, the more insight do the scientists gain and they thus continuously increase the detail of the meshes and build them as detailed as their instruments and the simulation hardware allow. In the process, the data volume also increases, slowing down the execution of spatial range queries needed to monitor the simulation considerably. Indexing speeds up range query execution, but the overhead to maintain the indexes is considerable because almost the entire mesh changes unpredictably at every simulation step. Using a simple linear scan, on the other hand, requires accessing the entire mesh and the performance deteriorates as the size of the dataset grows. In this paper we propose OCTOPUS, a strategy for executing range queries on mesh datasets that change unpredictably during simulations. In OCTOPUS we use the key insight that the mesh surface along with the mesh connectivity is sufficient to retrieve accurate query results efficiently. With this novel query execution strategy, OCTOPUS minimizes index maintenance cost and reduces query execution time considerably. Our experiments show that OCTOPUS achieves a speedup between 7.3 and 9.2× compared to the state of the art and that it scales better with increasing mesh dataset size and detail.

Computational Neuroscience Breakthroughs through Innovative Data Management

Simulations have become key in many scientific disciplines to better understand natural phenomena. Neuroscientists, for example, build and simulate increasingly fine-grained models including subcellular details, e.g., neurotransmitter of the neocortex to understand the mechanisms causing brain diseases and to test new treatments in-silico. The sheer size and, more importantly, the level of detail of their models challenges todays spatial data management techniques. In collaboration with the Blue Brain project BBP we develop new approaches that efficiently enable analysis, navigation and discovery in spatial models of the brain. More precisely, we develop an index for the scalable and efficient execution of spatial range queries supporting model building and analysis. Furthermore, we enable navigational access to the brain models, i.e., the execution of of series of range queries where he location of each query depends on the previous ones. To efficiently support navigational access, we develop a method that uses previous query results to prefetch spatial data with high accuracy and therefore speeds up navigation. Finally, to enable discovery based on the range queries, we conceive a novel in-memory spatial join. The methods we develop considerably outperform the state of the art, but more importantly, they enable the neuroscientists to scale to building, simulating and analyzing massively bigger and more detailed brain models.