arXiv subjects

David Patterson

Publications and source records attributed to David Patterson.

At least 37 records · Page 2Linked to original sources

arXiv2018

Enhanced spin coherence of rubidium atoms in solid parahydrogen

We measure the transverse relaxation of the spin state of an ensemble of ground-state rubidium atoms trapped in solid parahydrogen at cryogenic temperatures. We find the spin dephasing time of the ensemble (T$_2^*$) is limited by inhomogeneous broadening. We determine that this broadening is dominated by electrostatic interactions with the host matrix, and can be reduced by preparing nonclassical spin superposition states. Driving these superposition states gives significantly narrower electron paramagnetic resonance lines and the longest reported electron spin $T_2^*$ in any solid-phase system other than solid helium.

physics.atom-ph↗

arXiv2017

A Method for Preparation and Readout of Polyatomic Molecules in Single Quantum States

Polyatomic molecular ions contain many desirable attributes of a useful quantum system, including rich internal degrees of freedom and highly controllable coupling to the environment. To date, the vast majority of state-specific experimental work on molecular ions has concentrated on diatomic species. The ability to prepare and readout polyatomic molecules in single quantum states would enable diverse experimental avenues not available with diatomics, including new applications in precision measurement, sensitive chemical and chiral analysis at the single molecule level, and precise studies of Hz-level molecular tunneling dynamics. While cooling the motional state of a polyatomic ion via sympathetic cooling with a laser cooled atomic ion is straightforward, coupling this motional state to the internal state of the molecule has proven challenging. Here we propose a new method for readout and projective measurement of the internal state of a trapped polyatomic ion. The method exploits the rich manifold of technically accessible rotational states in the molecule to realize robust state-preparation and readout with far less stringent engineering than quantum logic methods recently demonstrated on diatomic molecules. The method can be applied to any reasonably small (less than about 10 atoms) polyatomic ion with an anisotropic polarizability.

quant-ph↗

arXiv2017

A Berkeley View of Systems Challenges for AI

With the increasing commoditization of computer vision, speech recognition and machine translation systems and the widespread deployment of learning-based back-end technologies such as digital advertising and intelligent infrastructures, AI (Artificial Intelligence) has moved from research labs to production. These changes have been made possible by unprecedented levels of data and computation, by methodological advances in machine learning, by innovations in systems software and architectures, and by the broad accessibility of these technologies. The next generation of AI systems promises to accelerate these developments and increasingly impact our lives via frequent interactions and making (often mission-critical) decisions on our behalf, often in highly personalized contexts. Realizing this promise, however, raises daunting challenges. In particular, we need AI systems that make timely and safe decisions in unpredictable environments, that are robust against sophisticated adversaries, and that can process ever increasing amounts of data across organizations and individuals without compromising confidentiality. These challenges will be exacerbated by the end of the Moore's Law, which will constrain the amount of data these technologies can store and process. In this paper, we propose several open research directions in systems, architectures, and security that can address these challenges and help unlock AI's potential to improve lives and society.

cs.AI↗

arXiv2017

The GAP Benchmark Suite

We present a graph processing benchmark suite with the goal of helping to standardize graph processing evaluations. Fewer differences between graph processing evaluations will make it easier to compare different research efforts and quantify improvements. The benchmark not only specifies graph kernels, input graphs, and evaluation methodologies, but it also provides optimized baseline implementations. These baseline implementations are representative of state-of-the-art performance, and thus new contributions should outperform them to demonstrate an improvement. The input graphs are sized appropriately for shared memory platforms, but any implementation on any platform that conforms to the benchmark's specifications could be compared. This benchmark suite can be used in a variety of settings. Graph framework developers can demonstrate the generality of their programming model by implementing all of the benchmark's kernels and delivering competitive performance on all of the benchmark's graphs. Algorithm designers can use the input graphs and the baseline implementations to demonstrate their contribution. Platform designers and performance analysts can use the suite as a workload representative of graph processing.

cs.DC↗

arXiv2017

Distributed-Memory Breadth-First Search on Massive Graphs

This chapter studies the problem of traversing large graphs using the breadth-first search order on distributed-memory supercomputers. We consider both the traditional level-synchronous top-down algorithm as well as the recently discovered direction optimizing algorithm. We analyze the performance and scalability trade-offs in using different local data structures such as CSR and DCSC, enabling in-node multithreading, and graph decompositions such as 1D and 2D decomposition.

cs.DC↗

arXiv2017

In-Datacenter Performance Analysis of a Tensor Processing Unit

Many architects believe that major improvements in cost-energy-performance must now come from domain-specific hardware. This paper evaluates a custom ASIC---called a Tensor Processing Unit (TPU)---deployed in datacenters since 2015 that accelerates the inference phase of neural networks (NN). The heart of the TPU is a 65,536 8-bit MAC matrix multiply unit that offers a peak throughput of 92 TeraOps/second (TOPS) and a large (28 MiB) software-managed on-chip memory. The TPU's deterministic execution model is a better match to the 99th-percentile response-time requirement of our NN applications than are the time-varying optimizations of CPUs and GPUs (caches, out-of-order execution, multithreading, multiprocessing, prefetching, ...) that help average throughput more than guaranteed latency. The lack of such features helps explain why, despite having myriad MACs and a big memory, the TPU is relatively small and low power. We compare the TPU to a server-class Intel Haswell CPU and an Nvidia K80 GPU, which are contemporaries deployed in the same datacenters. Our workload, written in the high-level TensorFlow framework, uses production NN applications (MLPs, CNNs, and LSTMs) that represent 95% of our datacenters' NN inference demand. Despite low utilization for some applications, the TPU is on average about 15X - 30X faster than its contemporary GPU or CPU, with TOPS/Watt about 30X - 80X higher. Moreover, using the GPU's GDDR5 memory in the TPU would triple achieved TOPS and raise TOPS/Watt to nearly 70X the GPU and 200X the CPU.

cs.AR↗

arXiv2017

Enantiomer-Specific State Transfer of Chiral Molecules

State-selective enantiomeric excess is realized using microwave-driven coherent population transfer. The method selectively promotes either R or S molecules to a higher rotational state by phase-controlled microwave pulses that drive electric-dipole allowed rotational transitions. We demonstrate the enantiomer-specific state transfer method using enantiopure samples of 1,2-propanediol. This method of state-specific enantiomeric enrichment can be applied to a large class of asymmetric, chiral molecules that can be vaporized and cooled to the point where rotationally resolved spectroscopy is possible, including molecules that rapidly racemize. The rapid chiral switching demonstrated here allows for new approaches in high-precision spectroscopic searches for parity violation in chiral molecules.

physics.chem-ph↗

arXiv2016

Continuous probe of cold complex molecules with infrared frequency comb spectroscopy

Cavity-enhanced frequency comb spectroscopy for molecule detection in the mid-infrared powerfully combines high resolution, high sensitivity, and broad spectral coverage. However, this technique, and essentially all spectroscopic methods, is limited in application to relatively small, simple molecules. Here we integrate comb spectroscopy with continuous, cold samples of molecules produced via buffer gas cooling, thus enabling the study of significantly more complex molecules. We report simultaneous gains in resolution, sensitivity, and bandwidth and demonstrate this combined capability with the first rotationally resolved direct absorption spectra in the CH stretch region of several complex molecules. These include nitromethane (CH$_3$NO$_2$), a model system that presents challenging questions to the understanding of large amplitude vibrational motion, as well as several large organic molecules with fundamental spectroscopic and astrochemical relevance, including naphthalene (C$_{10}$H$_8$), adamantane (C$_{10}$H$_{16}$), and hexamethylenetetramine (C$_{6}$N$_4$H$_{12}$). This general spectroscopic tool has the potential to significantly impact the field of molecular spectroscopy, simultaneously improving efficiency, spectral resolution, and specificity by orders of magnitude. This realization could open up new molecular species and new kinetics for precise investigations, including the study of complex molecules, weakly bound clusters, and cold chemistry.

physics.chem-ph↗

arXiv2016

Direct Time-domain Observation of Conformational Relaxation in Gas-phase Cold Collisions

Cooling molecules in the gas phase is important for precision spectroscopy, cold molecule physics, and physical chemistry. Measurements of conformational relaxation cross sections shed important light on potential energy surfaces and energy flow within a molecule. However, gas-phase conformational cooling has not been previously observed directly. In this work, we directly observe conformational dynamics of 1,2-propanediol in cold (6K) collisions with atomic helium using microwave spectroscopy and buffer-gas cooling. Precise knowledge and control of the collisional environment in the buffer-gas allows us to measure the absolute collision cross-section for conformational relaxation. Several conformers of 1,2-propanediol are investigated and found to have relaxation cross-sections with He ranging from $σ=4.7(3.0)\times10^{-18}\:\mathrm{cm}^{2}$ to $σ>5\times10^{-16}\:\mathrm{cm}^{2}$. Our method is applicable to a broad class of molecules and could be used to provide information about the potential energy surfaces of previously uninvestigated molecules.

physics.chem-ph↗

arXiv2014

Slow, Continuous Beams of Large Gas Phase Molecules

Cold, continuous, high flux beams of benzonitrile, fluorobenzine, and anisole have been created. Buffer-gas cooling with a cryogenic gas provides the cooling and slow forward beam velocities. The beam of benzonitrile was measured to have a forward velocity peaked at 67 $\pm 5$ m s$^{-1}$, and a continuous flux of $10^{15}$ molecules s$^{-1}$. These beams provide a continuous source for high resolution spectroscopy, and provide an attractive starting point for further spatial manipulation of such molecules, including eventual trapping.

physics.chem-ph↗

arXiv2014

SMaSH: A Benchmarking Toolkit for Human Genome Variant Calling

Motivation: Computational methods are essential to extract actionable information from raw sequencing data, and to thus fulfill the promise of next-generation sequencing technology. Unfortunately, computational tools developed to call variants from human sequencing data disagree on many of their predictions, and current methods to evaluate accuracy and computational performance are ad-hoc and incomplete. Agreement on benchmarking variant calling methods would stimulate development of genomic processing tools and facilitate communication among researchers. Results: We propose SMaSH, a benchmarking methodology for evaluating human genome variant calling algorithms. We generate synthetic datasets, organize and interpret a wide range of existing benchmarking data for real genomes, and propose a set of accuracy and computational performance metrics for evaluating variant calling methods on this benchmarking data. Moreover, we illustrate the utility of SMaSH to evaluate the performance of some leading single nucleotide polymorphism (SNP), indel, and structural variant calling algorithms. Availability: We provide free and open access online to the SMaSH toolkit, along with detailed documentation, at smash.cs.berkeley.edu.

q-bio.GN↗

arXiv2013

Sensitive Chiral Analysis via Microwave Three-wave Mixing

We demonstrate chirality-induced three-wave mixing in the microwave regime, using rotational transitions in cold gas-phase samples of 1,2-propanediol and 1,3-butanediol. We show that bulk three-wave mixing, which can only be realized in a chiral environment, provides a sensitive, species selective probe of enantiomeric excess and is applicable to a broad class of molecules. The doubly resonant condition provides simultaneous identification of species and of handedness, which should allow sensitive chiral analysis even within a complex mixture.

physics.chem-ph↗

arXiv2011

Faster and More Accurate Sequence Alignment with SNAP

We present the Scalable Nucleotide Alignment Program (SNAP), a new short and long read aligner that is both more accurate (i.e., aligns more reads with fewer errors) and 10-100x faster than state-of-the-art tools such as BWA. Unlike recent aligners based on the Burrows-Wheeler transform, SNAP uses a simple hash index of short seed sequences from the genome, similar to BLAST's. However, SNAP greatly reduces the number and cost of local alignment checks performed through several measures: it uses longer seeds to reduce the false positive locations considered, leverages larger memory capacities to speed index lookup, and excludes most candidate locations without fully computing their edit distance to the read. The result is an algorithm that scales well for reads from one hundred to thousands of bases long and provides a rich error model that can match classes of mutations (e.g., longer indels) that today's fast aligners ignore. We calculate that SNAP can align a dataset with 30x coverage of a human genome in less than an hour for a cost of $2 on Amazon EC2, with higher accuracy than BWA. Finally, we describe ongoing work to further improve SNAP.

cs.DS↗

arXiv2010

Search for the electric dipole moment of the electron with thorium monoxide

The electric dipole moment of the electron (eEDM) is a signature of CP-violating physics beyond the Standard Model. We describe an ongoing experiment to measure or set improved limits to the eEDM, using a cold beam of thorium monoxide (ThO) molecules. The metastable $H {}^3Δ_1$ state in ThO has important advantages for such an experiment. We argue that the statistical uncertainty of an eEDM measurement could be improved by as much as 3 orders of magnitude compared to the current experimental limit, in a first-generation apparatus using a cold ThO beam. We describe our measurements of the $H$ state lifetime and the production of ThO molecules in a beam, which provide crucial data for the eEDM sensitivity estimate. ThO also has ideal properties for the rejection of a number of known systematic errors; these properties and their implications are described.

physics.atom-ph↗

arXiv2010

Cold heteromolecular dipolar collisions

We present the first experimental observation of cold collisions between two different species of neutral polar molecules, each prepared in a single internal quantum state. Combining for the first time the techniques of Stark deceleration, magnetic trapping, and cryogenic buffer gas cooling allows the enhancement of molecular interaction time by 10$^5$. This has enabled an absolute measurement of the total trap loss cross sections between OH and ND$_3$ at a mean collision energy of 3.6 cm$^{-1}$ (5 K). Due to the dipolar interaction, the total cross section increases upon application of an external polarizing electric field. Cross sections computed from \emph{ab initio} potential energy surfaces are in excellent agreement with the measured value at zero external electric field. The theory presented here represents the first such analysis of collisions between a $^2Π$ radical and a closed-shell polyatomic molecule.

physics.chem-ph↗

arXiv2009

SCADS: Scale-Independent Storage for Social Computing Applications

Collaborative web applications such as Facebook, Flickr and Yelp present new challenges for storing and querying large amounts of data. As users and developers are focused more on performance than single copy consistency or the ability to perform ad-hoc queries, there exists an opportunity for a highly-scalable system tailored specifically for relaxed consistency and pre-computed queries. The Web 2.0 development model demands the ability to both rapidly deploy new features and automatically scale with the number of users. There have been many successful distributed key-value stores, but so far none provide as rich a query language as SQL. We propose a new architecture, SCADS, that allows the developer to declaratively state application specific consistency requirements, takes advantage of utility computing to provide cost effective scale-up and scale-down, and will use machine learning models to introspectively anticipate performance problems and predict the resource requirements of new queries before execution.

cs.DB↗

arXiv2009

Realization of Coherent Optically Dense Media via Buffer-Gas Cooling

We demonstrate that buffer-gas cooling combined with laser ablation can be used to create coherent optical media with high optical depth and low Doppler broadening that offers metastable states with low collisional and motional decoherence. Demonstration of this generic technique opens pathways to coherent optics with a large variety of atoms and molecules. We use helium buffer gas to cool 87Rb atoms to below 7 K and slow atom diffusion to the walls. Electromagnetically induced transparency (EIT) in this medium allows for 50% transmission in a medium with initial OD >70 and for slow pulse propagation with large delay-bandwidth products. In the high-OD regime, we observe high-contrast spectrum oscillations due to efficient four-wave mixing.

quant-ph↗

arXiv2008

Intense Atomic and Molecular Beams via Neon Buffer Gas Cooling

We realize a continuous guided beam of cold deuterated ammonia with a flux of 3e11 ND3 molecules/s and a continuous free-space beam of cold potassium with a flux of 1e16 K atoms/s. A novel feature of the buffer gas source used to produce these beams is cold neon, which, due to intermediate Knudsen number beam dynamics, produces a forward velocity and low-energy tail that is comparable to much colder helium-based sources. We expect this source to be trivially generalizable to a very wide range of atomic and molecular species with significant vapor pressure below 1000 K. This source has properties that make it a good starting point for laser cooling of molecules or atoms, cold collision studies, trapping, or nonlinear optics in buffer-gas-cooled atomic or molecular gases.

physics.atom-ph↗