arXiv subjects

Silu Huang

Publications and source records attributed to Silu Huang.

18 recordsLinked to original sources

arXiv2025

Cloud-Native Vector Search: A Comprehensive Performance Analysis

Vector search has been widely employed in recommender system and retrieval-augmented-generation pipelines, commonly performed with vector indexes to efficiently find similar items in large datasets. Recent growths in both data and task complexity have motivated placing vector indexes onto remote storage -- cloud-native vector search, which cloud providers have recently introduced services for. Yet, despite varying workload characteristics and various available vector index forms, providers default to using cluster-based indexes, which on paper do adapt well to differences between disk and cloud-based environment: their fetch granularities and lack of notable intra-query dependencies aligns with the large optimal fetch sizes and minimizes costly round-trips (i.e., as opposed to graph-based indexes) to remote storage, respectively. This paper systematically studies cloud-native vector search: What and how should indexes be built and used for on-cloud vector search? We analyze bottlenecks of two common index classes, cluster and graph indexes, on remote storage, and show that despite current standardized adoption of cluster indexes on the cloud, graph indexes are favored in workloads requiring high concurrency and recall, or operating on high-dimensional data or large datatypes. We further find that on-cloud search demands significantly different indexing and search parameterizations versus on-disk search for optimal performance. Finally, we incorporate existing cloud-based caching setups into vector search and find that certain index optimizations work against caching, and study how this can be mitigated to maximize gains under various available cache sizes.

cs.DB↗

arXiv2025

SIEVE: Effective Filtered Vector Search with Collection of Indexes

Many real-world tasks such as recommending videos with the kids tag can be reduced to finding most similar vectors associated with hard predicates. This task, filtered vector search, is challenging as prior state-of-the-art graph-based (unfiltered) similarity search techniques quickly degenerate when hard constraints are considered. That is, effective graph-based filtered similarity search relies on sufficient connectivity for reaching the most similar items within just a few hops. To consider predicates, recent works propose modifying graph traversal to visit only the items that may satisfy predicates. However, they fail to offer the just-a-few-hops property for a wide range of predicates: they must restrict predicates significantly or lose efficiency if only a small fraction of items satisfy predicates. We propose an opposite approach: instead of constraining traversal, we build many indexes each serving different predicate forms. For effective construction, we devise a three-dimensional analytical model capturing relationships among index size, search time, and recall, with which we follow a workload-aware approach to pack as many useful indexes as possible into a collection. At query time, the analytical model is employed yet again to discern the one that offers the fastest search at a given recall. We show superior performance and support on datasets with varying selectivities and forms: our approach achieves up to 8.06x speedup while having as low as 1% build time versus other indexes, with less than 2.15x memory of a standard HNSW graph and modest knowledge of past workloads.

cs.DB↗

arXiv2024

Mixed Precision Block-Jacobi Preconditioner: Algorithms, Performance Evaluation and Feature Analysis

In this paper, we propose two mixed precision algorithms for Block-Jacobi preconditioner(BJAC): a fixed low precision strategy and an adaptive precision strategy. We evaluate the performance improvement of the proposed mixed precision BJAC preconditioners combined with the preconditioned conjugate gradient algorithm using problems including diffusion equations and radiation hydrodynamics equations. Numerical results show that, compared to the uniform high precision PCG algorithm, the mixed precision preconditioners can achieve speedups from 1.3 to 1.8 without sacrificing accuracy. Furthermore, we observe the phenomenon of convergence delay in some test cases for the mixed precision preconditioners, and further analyse the matrix features associate with the convergence delay behavior.

math.NA↗

arXiv2023

Multigap nodeless superconductivity in the topological semimetal PdTe

Recently PdTe was identified as a spin-orbit coupled topological Dirac semimetal and was claimed to exhibit both bulk-nodal and surface-nodeless superconducting gaps. Here we report the ultralow-temperature thermal conductivity measurements on PdTe single crystals with $T_c$ = 4.5 K to investigate its superconducting gap structure. It is found that the residual linear term $κ_0/T$ is negligible in zero magnetic field. Furthermore, the field dependence of $κ_0(H)/T$ exhibits an $\sf S$-shaped curve. These results suggest that PdTe has multiple nodeless superconducting gaps, which is at odds with the claimed bulk-nodal gap. The reason for the discrepancy is likely that previous angle-resolved photoemission spectroscopy measurements were only performed down to 2 K and cannot observe the smaller nodeless gap. The fully gapped superconducting state in PdTe is compatible with it being a topological superconductor candidate.

cond-mat.supr-con↗

arXiv2023

Convergence estimation and characteristic analysis of a two-level iterative algorithm for the discretized three-temperature energy linear systems

For solving the discretized three-temperature energy linear systems, Xu et al. proposed a physical-variable based coarsening two-level iterative method (PCTL algorithm) in 2009 and verified its efficiency by numerical experiments in practical applications. In this paper, we study in detail the specific convergence property of the PCTL algorithm based on the theory of algebraic multigrid method (AMG),and give a reasonable upper bound on the convergence factor, which provides a theoretical guarantee for the PCTL algorithm. Moreover, we also analyse the algebraic features that affect the convergence of the PCTL algorithm, such as diagonal dominance and coupling strength, hoping provides theoretical guidance for the applications and algorithm optimization of the PCTL algorithm.

math.NA↗

arXiv2022

TSEXPLAIN: Explaining Aggregated Time Series by Surfacing Evolving Contributors

Aggregated time series are generated effortlessly everywhere, e.g., "total confirmed covid-19 cases since 2019" and "total liquor sales over time." Understanding "how" and "why" these key performance indicators (KPI) evolve over time is critical to making data-informed decisions. Existing explanation engines focus on explaining one aggregated value or the difference between two relations. However, this falls short of explaining KPIs' continuous changes over time. Motivated by this, we propose TSEXPLAIN, a system that explains aggregated time series by surfacing the underlying evolving top contributors. Under the hood, we leverage prior works on two-relations diff as a building block and formulate a K-Segmentation problem to segment the time series such that each segment after segmentation shares consistent explanations, i.e., contributors. To quantify consistency in each segment, we propose a novel within-segment variance design that is explanation-aware; to derive the optimal K-Segmentation scheme, we develop an efficient dynamic programming algorithm. Experiments on synthetic and real-world datasets show that our explanation-aware segmentation can effectively identify evolving explanations for aggregated time series and outperform explanation-agnostic segmentation. Further, we proposed an optimal selection strategy of K and several optimizations to speed up TSEXPLAIN for interactive user experience, achieving up to 13X efficiency improvement.

cs.DB↗

arXiv2020

Frugal Optimization for Cost-related Hyperparameters

The increasing demand for democratizing machine learning algorithms calls for hyperparameter optimization (HPO) solutions at low cost. Many machine learning algorithms have hyperparameters which can cause a large variation in the training cost. But this effect is largely ignored in existing HPO methods, which are incapable to properly control cost during the optimization process. To address this problem, we develop a new cost-frugal HPO solution. The core of our solution is a simple but new randomized direct-search method, for which we prove a convergence rate of $O(\frac{\sqrt{d}}{\sqrt{K}})$ and an $O(dε^{-2})$-approximation guarantee on the total cost. We provide strong empirical results in comparison with state-of-the-art HPO methods on large AutoML benchmarks.

cs.LG↗

arXiv2018

ABC: Efficient Selection of Machine Learning Configuration on Large Dataset

A machine learning configuration refers to a combination of preprocessor, learner, and hyperparameters. Given a set of configurations and a large dataset randomly split into training and testing set, we study how to efficiently select the best configuration with approximately the highest testing accuracy when trained from the training set. To guarantee small accuracy loss, we develop a solution using confidence interval (CI)-based progressive sampling and pruning strategy. Compared to using full data to find the exact best configuration, our solution achieves more than two orders of magnitude speedup, while the returned top configuration has identical or close test accuracy.

cs.LG↗

arXiv2018

Adaptive Sampling for Rapidly Matching Histograms

In exploratory data analysis, analysts often have a need to identify histograms that possess a specific distribution, among a large class of candidate histograms, e.g., find countries whose income distribution is most similar to that of Greece. This distribution could be a new one that the user is curious about, or a known distribution from an existing histogram visualization. At present, this process of identification is brute-force, requiring the manual generation and evaluation of a large number of histograms. We present FastMatch: an end-to-end approach for interactively retrieving the histogram visualizations most similar to a user-specified target, from a large collection of histograms. The primary technical contribution underlying FastMatch is a probabilistic algorithm, HistSim, a theoretically sound sampling-based approach to identify the top-$k$ closest histograms under $\ell_1$ distance. While HistSim can be used independently, within FastMatch we couple HistSim with a novel system architecture that is aware of practical considerations, employing asynchronous block-based sampling policies, building on lightweight sampling engines developed in recent work. FastMatch obtains near-perfect accuracy with up to $35\times$ speedup over approaches that do not use sampling on several real-world datasets.

cs.DB↗

arXiv2018

Navigating the Data Lake with Datamaran: Automatically Extracting Structure from Log Datasets

Organizations routinely accumulate semi-structured log datasets generated as the output of code; these datasets remain unused and uninterpreted, and occupy wasted space - this phenomenon has been colloquially referred to as "data lake" problem. One approach to leverage these semi-structured datasets is to convert them into a structured relational format, following which they can be analyzed in conjunction with other datasets. We present Datamaran, an tool that extracts structure from semi-structured log datasets with no human supervision. Datamaran automatically identifies field and record endpoints, separates the structured parts from the unstructured noise or formatting, and can tease apart multiple structures from within a dataset, in order to efficiently extract structured relational datasets from semi-structured log datasets, at scale with high accuracy. Compared to other unsupervised log dataset extraction tools developed in prior work, Datamaran does not require the record boundaries to be known beforehand, making it much more applicable to the noisy log files that are ubiquitous in data lakes. Datamaran can successfully extract structured information from all datasets used in prior work, and can achieve 95% extraction accuracy on automatically collected log datasets from GitHub - a substantial 66% increase of accuracy compared to unsupervised schemes from prior work. Our user study further demonstrates that the extraction results of Datamaran are closer to the desired structure than competing algorithms.

cs.DB↗

arXiv2017

Optimally Leveraging Density and Locality to Support LIMIT Queries

Existing database systems are not optimized for queries with a LIMIT clause---operating instead in an all-or-nothing manner. In this paper, we propose a fast LIMIT query evaluation engine, called NeedleTail, aimed at letting analysts browse a small sample of the query results on large datasets as quickly as possible, independent of the overall size of the result set. NeedleTail introduces density maps, a lightweight in-memory indexing structure, and a set of efficient algorithms (with desirable theoretical guarantees) to quickly locate promising blocks, trading off locality and density. In settings where the samples are used to compute aggregates, we extend techniques from survey sampling to mitigate the bias in our samples. Our experimental results demonstrate that NeedleTail returns results 4x faster on HDDs and 9x faster on SSDs on average, while occupying up to 23x less memory than existing techniques.

cs.DB↗

arXiv2017

Nontrivial Berry phase in magnetic BaMnSb2 semimetal

The subject of topological materials has attracted immense attention in condensed-matter physics, because they host new quantum states of matter containing Dirac, Majorana, or Weyl fermions. Although Majorana fermions can only exist on the surface of topological superconductors, Dirac and Weyl fermions can be realized in both 2D and 3D materials. The latter are semimetals with Dirac/Weyl cones either not tilted (type I) or tilted (type II). Although both Dirac and Weyl fermions have massless nature with the nontrivial Berry phase, the formation of Weyl fermions in 3D semimetals require either time-reversal or inversion symmetry breaking to lift degeneracy at Dirac points. Here, we demonstrate experimentally that canted antiferromagnetic BaMnSb2 is a 3D Weyl semimetal with a 2D electronic structure. The Shubnikov-de Hass oscillations of the magnetoresistance give nearly zero effective mass with high mobility and the nontrivial Berry phase. The ordered magnetic arrangement (ferromagnetic ordering in the ab plane and antiferromagnetic ordering along the c axis below 286 K) breaks the time-reversal symmetry, thus offering us an ideal platform to study magnetic Weyl fermions in a centrosymmetric material.

cond-mat.mtrl-sci↗

arXiv2017

OrpheusDB: Bolt-on Versioning for Relational Databases

Data science teams often collaboratively analyze datasets, generating dataset versions at each stage of iterative exploration and analysis. There is a pressing need for a system that can support dataset versioning, enabling such teams to efficiently store, track, and query across dataset versions. We introduce OrpheusDB, a dataset version control system that "bolts on" versioning capabilities to a traditional relational database system, thereby gaining the analytics capabilities of the database "for free". We develop and evaluate multiple data models for representing versioned data, as well as a light-weight partitioning scheme, LyreSplit, to further optimize the models for reduced query latencies. With LyreSplit, OrpheusDB is on average 1000x faster in finding effective (and better) partitionings than competing approaches, while also reducing the latency of version retrieval by up to 20x relative to schemes without partitioning. LyreSplit can be applied in an online fashion as new versions are added, alongside an intelligent migration scheme that reduces migration time by 10x on average.

cs.DB↗

arXiv2016

Finding Multiple New Optimal Locations in a Road Network

We study the problem of optimal location querying for location based services in road networks, which aims to find locations for new servers or facilities. The existing optimal solutions on this problem consider only the cases with one new server. When two or more new servers are to be set up, the problem with minmax cost criteria, MinMax, becomes NP-hard. In this work we identify some useful properties about the potential locations for the new servers, from which we derive a novel algorithm for MinMax, and show that it is efficient when the number of new servers is small. When the number of new servers is large, we propose an efficient 3-approximate algorithm. We verify with experiments on real road networks that our solutions are effective and attains significantly better result quality compared to the existing greedy algorithms.

cs.DB↗

arXiv2015

Towards a unified query language for provenance and versioning

Organizations and teams collect and acquire data from various sources, such as social interactions, financial transactions, sensor data, and genome sequencers. Different teams in an organization as well as different data scientists within a team are interested in extracting a variety of insights which require combining and collaboratively analyzing datasets in diverse ways. DataHub is a system that aims to provide robust version control and provenance management for such a scenario. To be truly useful for collaborative data science, one also needs the ability to specify queries and analysis tasks over the versioning and the provenance information in a unified manner. In this paper, we present an initial design of our query language, called VQuel, that aims to support such unified querying over both types of information, as well as the intermediate and final results of analyses. We also discuss some of the key language design and implementation challenges moving forward.

cs.DB↗

arXiv2015

Principles of Dataset Versioning: Exploring the Recreation/Storage Tradeoff

The relative ease of collaborative data science and analysis has led to a proliferation of many thousands or millions of $versions$ of the same datasets in many scientific and commercial domains, acquired or constructed at various stages of data analysis across many users, and often over long periods of time. Managing, storing, and recreating these dataset versions is a non-trivial task. The fundamental challenge here is the $storage-recreation\;trade-off$: the more storage we use, the faster it is to recreate or retrieve versions, while the less storage we use, the slower it is to recreate or retrieve versions. Despite the fundamental nature of this problem, there has been a surprisingly little amount of work on it. In this paper, we study this trade-off in a principled manner: we formulate six problems under various settings, trading off these quantities in various ways, demonstrate that most of the problems are intractable, and propose a suite of inexpensive heuristics drawing from techniques in delay-constrained scheduling, and spanning tree literature, to solve these problems. We have built a prototype version management system, that aims to serve as a foundation to our DATAHUB system for facilitating collaborative data science. We demonstrate, via extensive experiments, that our proposed heuristics provide efficient solutions in practical dataset versioning scenarios.

cs.DB↗

arXiv2014

(α, k)-Minimal Sorting and Skew Join in MPI and MapReduce

As computer clusters are found to be highly effective for handling massive datasets, the design of efficient parallel algorithms for such a computing model is of great interest. We consider (α, k)-minimal algorithms for such a purpose, where α is the number of rounds in the algorithm, and k is a bound on the deviation from perfect workload balance. We focus on new (α, k)-minimal algorithms for sorting and skew equijoin operations for computer clusters. To the best of our knowledge the proposed sorting and skew join algorithms achieve the best workload balancing guarantee when compared to previous works. Our empirical study shows that they are close to optimal in workload balancing. In particular, our proposed sorting algorithm is around 25% more efficient than the state-of-the-art Terasort algorithm and achieves significantly more even workload distribution by over 50%.

cs.DB↗

arXiv2014

Temporal Graph Traversals: Definitions, Algorithms, and Applications

A temporal graph is a graph in which connections between vertices are active at specific times, and such temporal information leads to completely new patterns and knowledge that are not present in a non-temporal graph. In this paper, we study traversal problems in a temporal graph. Graph traversals, such as DFS and BFS, are basic operations for processing and studying a graph. While both DFS and BFS are well-known simple concepts, it is non-trivial to adopt the same notions from a non-temporal graph to a temporal graph. We analyze the difficulties of defining temporal graph traversals and propose new definitions of DFS and BFS for a temporal graph. We investigate the properties of temporal DFS and BFS, and propose efficient algorithms with optimal complexity. In particular, we also study important applications of temporal DFS and BFS. We verify the efficiency and importance of our graph traversal algorithms in real world temporal graphs.

cs.DS↗