arXiv subjects

Yongluan Zhou

Publications and source records attributed to Yongluan Zhou.

At least 19 recordsLinked to original sources

arXiv2026

vMODB: Unifying Event and Data Management for Distributed Asynchronous Applications

Event-driven microservice architecture (EDMA) has emerged as a crucial architectural pattern for scalable cloud applications. In typical EDMAs, database systems are relegated to isolated storage engines for individual components, blind to cross-component transactions, while messaging systems are unaware of each component's application state. Consequently, EDMAs impose a severe trade-off: developers must either sacrifice strong data consistency and integrity or manually manage complex distributed coordination. To address this challenge, we design vMODB, a distributed framework that offers a better trade-off and enables developers to build highly consistent and scalable cloud applications without compromising the benefits of EDMA. The core contribution of vMODB lies in the co-design of a programming abstraction and the underlying specialized system. We propose Virtual Micro Service (VMS), a novel programming model that provides familiar Object-Relational Mapping (ORM) and meta-programming constructs for specifying the data model, constraints, concurrency, and dependencies, making application semantics visible to the system. vMODB leverages semantic visibility to enforce ACID properties by transparently unifying event logs and state management, relieving developers from the burden of ensuring cross-component data consistency and integrity. Thanks to full-stack system optimizations enabled by our co-design, experiments using two benchmarks show that vMODB outperforms a widely adopted state-of-the-art competing framework that only offers eventual consistency by up to 3x.

cs.DB↗

arXiv2025

Redox: Improving I/O Efficiency of Model Training Through File Redirection

This paper proposes Redox, a training data management system designed to achieve high I/O efficiency. The key insight is a new observation of file redirection: for model training, when training data in one file is requested, the system has the flexibility to return the data of another file. Based on this property, Redox starts with a bold design principle that chunks of data files are always read from disk in batch, and once loaded, all files in the chunk will be consumed without being loaded again. We propose efficient local and distributed file read protocol based on this principle that both minimizes the wasted data read and enables opportunistic prefetch from remote node. Moreover, we analyze file redirection's impact on randomness, and show that it has little effects on training efficiency. Experimental results indicate that Redox significantly accelerates data fetching in training, achieving up to a 4.57x improvement in end-to-end training compared to PyTorch.

cs.DC↗

arXiv2025

Dolphin: An Actor-Oriented Database for Reactive Moving Object Data Management

Novel reactive moving object applications require solutions to support object reactive behaviors as a way to query and update dynamic data. While moving object scenarios have long been researched in the context of spatio-temporal data management, reactive behavior is usually left to complex end-user implementations. However, it is not just a matter of hardwiring reactive constraints: the required solutions need to satisfy tight low-latency computation requirements and be scalable. This paper explores a novel approach to enrich a distributed actor-based framework with reactive functionality and complex spatial data management along with concurrency semantics. Our approach relies on a proposal of the moving actor abstraction, which is a conceptual enhancement of the actor model with reactive sensing, movement, and spatial querying capabilities. This enhancement helps developers of reactive moving object applications avoid the significant burden of implementing application-level schemes to balance performance and consistency. Based on moving actors, we define a reactive moving object data management platform, named Moving Actor-Oriented Databases (M-AODBs), and build Dolphin -- an implementation of M-AODBs. Dolphin embodies a non-intrusive actor-based design layered on top of the Microsoft Orleans distributed virtual actor framework. In a set of experimental evaluations with realistic reactive moving object scenarios, Dolphin exhibits scalability on multi-machines and provides near-real-time reaction latency.

cs.DB↗

arXiv2025

CFP: Efficient Optimization of Intra-Operator Parallelism Plans for Large Model Training

Optimizing the parallel training of large models requires exploring intra-operator parallelism plans for a computation graph that typically contains tens of thousands of primitive operators. While the optimization of parallel data processing graphs has been extensively researched in database systems, the vast search space makes it challenging to apply traditional database query optimization methods and algorithms. This paper introduces CFP, an optimization system for intra-operator parallelism that significantly reduces the complexity of searching for parallelism plans by leveraging two structural patterns found in large models. First, we identify parallel-preserving subgraphs, which ensure that the optimal global plan assigns the same parallel strategy to all operators within the subgraph. This approach allows us to avoid enumerating all possible combinations of parallel strategies for these operators. Second, we recognize repetitive subgraph patterns within the large computational graph, enabling us to profile a moderate number of representative subgraphs and accurately estimate the cost of parallelism plans with low overhead. With the significantly reduced search space, we can employ dynamic programming to search for the optimized parallelism plan. In our experiments, we demonstrate that CFP achieves significant speedups compared to the state-of-the-art framework for large models like GPT and LLAMA.

cs.DC↗

arXiv2025

Transactional Cloud Applications: Status Quo, Challenges, and Opportunities

Transactional cloud applications such as payment, booking, reservation systems, and complex business workflows are currently being rewritten for deployment in the cloud. This migration to the cloud is happening mainly for reasons of cost and scalability. Over the years, application developers have used different migration approaches, such as microservice frameworks, actors, and stateful dataflow systems. The migration to the cloud has brought back data management challenges traditionally handled by database management systems. Those challenges include ensuring state consistency, maintaining durability, and managing the application lifecycle. At the same time, the shift to a distributed computing infrastructure introduced new issues, such as message delivery, task scheduling, containerization, and (auto)scaling. Although the data management community has made progress in developing analytical and transactional database systems, transactional cloud applications have received little attention in database research. This tutorial aims to highlight recent trends in the area and discusses open research challenges for the data management community.

cs.DB↗

arXiv2025

Online Marketplace: A Benchmark for Data Management in Microservices

Microservice architectures have become a popular approach for designing scalable distributed applications. Despite their extensive use in industrial settings for over a decade, there is limited understanding of the data management challenges that arise in these applications. Consequently, it has been difficult to advance data system technologies that effectively support microservice applications. To fill this gap, we present Online Marketplace, a microservice benchmark that highlights core data management challenges that existing benchmarks fail to address. These challenges include transaction processing, query processing, event processing, constraint enforcement, and data replication. We have defined criteria for various data management issues to enable proper comparison across data systems and platforms. Through case studies with state-of-the-art data platforms, we discuss the issues encountered while implementing and meeting Online Marketplace's criteria. By capturing the overhead of meeting the key data management requirements that are overlooked by existing benchmarks, we gain actionable insights into the experimental platforms. This highlights the significance of Online Marketplace in advancing future data systems to meet the needs of microservice practitioners.

cs.DB↗

arXiv2024

Validated Strong Consensus Protocol for Asynchronous Vote-based Blockchains

Vote-based blockchains construct a state machine replication (SMR) system among participating nodes, using Byzantine Fault Tolerance (BFT) consensus protocols to transition from one state to another. Currently, they rely on either synchronous or partially synchronous networks with leader-based coordination or costly Asynchronous Common Subset (ACS) protocols in asynchronous settings, making them impractical for large-scale asynchronous applications. To make Asynchronous SMR scalable, this paper proposes a \emph{validated strong} BFT consensus model that allows leader-based coordination in asynchronous settings. Our BFT consensus model offers the same level of tolerance as binary byzantine agreement but does not demand consistency among honest nodes before they vote. An SMR using our model allows nodes to operate in different, tentative, but mutually exclusive states until they eventually converge on the same state. We propose an asynchronous BFT protocol for vote-based blockchains employing our consensus model to address several critical challenges: how to ensure that nodes eventually converge on the same state across voting rounds, how to assure that a blockchain will steadily progress through epochs while reaching consensus for previous epochs, and how to maintain robust byzantine fault tolerance. Our protocol greatly reduces message complexity and is the first one to achieve linear view changes without relying on threshold signatures. We prove that an asynchronous blockchain built on our protocol can operate with the \emph{same} simplicity and efficiency as partially synchronous blockchains built on, e.g. HotStuff-2. This facilitates deploying asynchronous blockchains across large-scale networks.

cs.DC↗

arXiv2024

gECC: A GPU-based high-throughput framework for Elliptic Curve Cryptography

Elliptic Curve Cryptography (ECC) is an encryption method that provides security comparable to traditional techniques like Rivest-Shamir-Adleman (RSA) but with lower computational complexity and smaller key sizes, making it a competitive option for applications such as blockchain, secure multi-party computation, and database security. However, the throughput of ECC is still hindered by the significant performance overhead associated with elliptic curve (EC) operations. This paper presents gECC, a versatile framework for ECC optimized for GPU architectures, specifically engineered to achieve high-throughput performance in EC operations. gECC incorporates batch-based execution of EC operations and microarchitecture-level optimization of modular arithmetic. It employs Montgomery's trick to enable batch EC computation and incorporates novel computation parallelization and memory management techniques to maximize the computation parallelism and minimize the access overhead of GPU global memory. Also, we analyze the primary bottleneck in modular multiplication by investigating how the user codes of modular multiplication are compiled into hardware instructions and what these instructions' issuance rates are. We identify that the efficiency of modular multiplication is highly dependent on the number of Integer Multiply-Add (IMAD) instructions. To eliminate this bottleneck, we propose techniques to minimize the number of IMAD instructions by leveraging predicate registers to pass the carry information and using addition and subtraction instructions (IADD3) to replace IMAD instructions. Our results show that, for ECDSA and ECDH, gECC can achieve performance improvements of 5.56x and 4.94x, respectively, compared to the state-of-the-art GPU-based system. In a real-world blockchain application, we can achieve performance improvements of 1.56x, compared to the state-of-the-art CPU-based system.

cs.CR↗

arXiv2024

Rethinking State Management in Actor Systems for Cloud-Native Applications

The actor model has gained increasing popularity. However, it lacks support for complex state management tasks, such as enforcing foreign key constraints and ensuring data replication consistency across actors. These are crucial properties in partitioned application designs, such as microservices. To fill this gap, we start by analyzing the key impediments in state-of-the-art actor systems. We find it difficult for developers to express complex data relationships across actors and reason about the impact of state updates on performance due to opaque state management abstractions. To solve this conundrum, we develop SmSa, a novel data management layer for actor systems, allowing developers to declare data dependencies that cut across actors, including foreign keys, data replications, and other dependencies. SmSa can transparently enforce the declared dependencies, reducing the burden on developers. Furthermore, SmSa employs novel logging and concurrency control algorithms to support transactional maintenance of data dependencies. We demonstrate SmSa can support core data management tasks where dependencies across components appear frequently without jeopardizing application logic expressiveness and performance. Our experiments show SmSa significantly reduces the logging overhead and leads to increased concurrency level, improving by up to 2X the performance of state-of-the-art deterministic scheduling approaches. As a result, SmSa will make it easier to design and implement highly partitioned and distributed applications.

cs.DB↗

arXiv2024

An Empirical Study on Challenges of Event Management in Microservice Architectures

Microservices emerged as a popular architectural style over the last decade. Although microservices are designed to be self-contained, they must communicate to realize business capabilities, creating dependencies among their data and functionalities. Developers then resort to asynchronous, event-based communication to fulfill such dependencies while reducing coupling. However, developers are often oblivious to the inherent challenges of the asynchronous and event-based paradigm, leading to frustrations and ultimately making them reconsider the adoption of microservices. To make matters worse, there is a scarcity of literature on the practices and challenges of designing, implementing, testing, monitoring, and troubleshooting event-based microservices. To fill this gap, this paper provides the first comprehensive characterization of event management practices and challenges in microservices based on a repository mining study of over 8000 Stack Overflow questions. Moreover, 628 relevant questions were randomly sampled for an in-depth manual investigation of challenges. We find that developers encounter many problems, including large event payloads, modeling event schemas, auditing event flows, and ordering constraints in processing events. This suggests that developers are not sufficiently served by state-of-the-practice technologies. We provide actionable implications to developers, technology providers, and researchers to advance event management in microservices.

cs.SE↗

arXiv2024

A Two-Layer Blockchain Sharding Protocol Leveraging Safety and Liveness for Enhanced Performance

Sharding is essential for improving blockchain scalability. Existing protocols overlook diverse adversarial attacks, limiting transaction throughput. This paper presents Reticulum, a groundbreaking sharding protocol addressing this issue, boosting blockchain scalability. Reticulum employs a two-phase approach, adapting transaction throughput based on runtime adversarial attacks. It comprises "control" and "process" shards in two layers. Process shards contain at least one trustworthy node, while control shards have a majority of trusted nodes. In the first phase, transactions are written to blocks and voted on by nodes in process shards. Unanimously accepted blocks are confirmed. In the second phase, blocks without unanimous acceptance are voted on by control shards. Blocks are accepted if the majority votes in favor, eliminating first-phase opponents and silent voters. Reticulum uses unanimous voting in the first phase, involving fewer nodes, enabling more parallel process shards. Control shards finalize decisions and resolve disputes. Experiments confirm Reticulum's innovative design, providing high transaction throughput and robustness against various network attacks, outperforming existing sharding protocols for blockchain networks.

cs.CR↗

arXiv2024

CLIP-Branches: Interactive Fine-Tuning for Text-Image Retrieval

The advent of text-image models, most notably CLIP, has significantly transformed the landscape of information retrieval. These models enable the fusion of various modalities, such as text and images. One significant outcome of CLIP is its capability to allow users to search for images using text as a query, as well as vice versa. This is achieved via a joint embedding of images and text data that can, for instance, be used to search for similar items. Despite efficient query processing techniques such as approximate nearest neighbor search, the results may lack precision and completeness. We introduce CLIP-Branches, a novel text-image search engine built upon the CLIP architecture. Our approach enhances traditional text-image search engines by incorporating an interactive fine-tuning phase, which allows the user to further concretize the search query by iteratively defining positive and negative examples. Our framework involves training a classification model given the additional user feedback and essentially outputs all positively classified instances of the entire data catalog. By building upon recent techniques, this inference phase, however, is not implemented by scanning the entire data catalog, but by employing efficient index structures pre-built for the data. Our results show that the fine-tuned results can improve the initial search outputs in terms of relevance and accuracy while maintaining swift response times

cs.IR↗

arXiv2024

Benchmarking Data Management Systems for Microservices

Microservice architectures are a popular choice for deploying large-scale data-intensive applications. This architectural style allows microservice practitioners to achieve requirements related to loose coupling, fault contention, workload isolation, higher data availability, scalability, and independent schema evolution. Although the industry has been employing microservices for over a decade, existing microservice benchmarks lack essential data management challenges observed in practice, including distributed transaction processing, consistent data querying and replication, event processing, and data integrity constraint enforcement. This gap jeopardizes the development of novel data systems that embrace the complex nature of data-intensive microservices. In this talk, we share our experience in designing Online Marketplace, a novel benchmark that embraces core data management requirements intrinsic to real-world microservices. By implementing the benchmark in state-of-the-art data platforms, we experience the pain practitioners face in assembling several heterogeneous components to realize their requirements. Our evaluation demonstrates Online Marketplace allows experimenting key properties sought by microservice practitioners, thus fomenting the design of novel data management systems.

cs.DB↗

arXiv2023

The Case of Transparent Cache Invalidation in Web Applications

Application-level caches are widely adopted by web applications to minimize the response time of user requests as well as to reduce the burden on the system backend, such as the database servers. In the state of practice, developers have to take care of the data freshness of application-level caches manually. Given the growing complexities of today's web applications, it becomes increasingly challenging for developers to understand, reason about, and implement cache invalidation methods. Furthermore, according to our survey of open-source web application projects and engineers, it is indeed challenging to map database updates with cache entries at the application level. Therefore, we propose a design to handle data validity in a transparent and precise manner, without requiring any intervention from developers. Its main idea is to modify the DBMS to provide necessary information for cache management and enhance the cache with an invalidation index to identify and invalidate outdated data automatically and efficiently. Based on the design, we further provide two specific solutions. Our preliminary experiments indicate that our solutions could effectively achieve transparent cache invalidation while maintaining cost-effectiveness.

cs.DB↗

arXiv2023

RapidEarth: A Search-by-Classification Engine for Large-Scale Geospatial Imagery

Data exploration and analysis in various domains often necessitate the search for specific objects in massive databases. A common search strategy, often known as search-by-classification, resorts to training machine learning models on small sets of positive and negative samples and to performing inference on the entire database to discover additional objects of interest. While such an approach often yields very good results in terms of classification performance, the entire database usually needs to be scanned, a process that can easily take several hours even for medium-sized data catalogs. In this work, we present RapidEarth, a geospatial search-by-classification engine that allows analysts to rapidly search for interesting objects in very large data collections of satellite imagery in a matter of seconds, without the need to scan the entire data catalog. RapidEarth embodies a co-design of multidimensional indexing structures and decision branches, a recently proposed variant of classical decision trees. These decision branches allow RapidEarth to transform the inference phase into a set of range queries, which can be efficiently processed by leveraging the aforementioned multidimensional indexing structures. The main contribution of this work is a geospatial search engine that implements these technical findings.

cs.DB↗

arXiv2023

Fast Search-By-Classification for Large-Scale Databases Using Index-Aware Decision Trees and Random Forests

The vast amounts of data collected in various domains pose great challenges to modern data exploration and analysis. To find "interesting" objects in large databases, users typically define a query using positive and negative example objects and train a classification model to identify the objects of interest in the entire data catalog. However, this approach requires a scan of all the data to apply the classification model to each instance in the data catalog, making this method prohibitively expensive to be employed in large-scale databases serving many users and queries interactively. In this work, we propose a novel framework for such search-by-classification scenarios that allows users to interactively search for target objects by specifying queries through a small set of positive and negative examples. Unlike previous approaches, our framework can rapidly answer such queries at low cost without scanning the entire database. Our framework is based on an index-aware construction scheme for decision trees and random forests that transforms the inference phase of these classification models into a set of range queries, which in turn can be efficiently executed by leveraging multidimensional indexing structures. Our experiments show that queries over large data catalogs with hundreds of millions of objects can be processed in a few seconds using a single server, compared to hours needed by classical scanning-based approaches.

cs.DB↗

arXiv2021

Data Management in Microservices: State of the Practice, Challenges, and Research Directions

Microservices have become a popular architectural style for data-driven applications, given their ability to functionally decompose an application into small and autonomous services to achieve scalability, strong isolation, and specialization of database systems to the workloads and data formats of each service. Despite the accelerating industrial adoption of this architectural style, an investigation of the state of the practice and challenges practitioners face regarding data management in microservices is lacking. To bridge this gap, we conducted a systematic literature review of representative articles reporting the adoption of microservices, we analyzed a set of popular open-source microservice applications, and we conducted an online survey to cross-validate the findings of the previous steps with the perceptions and experiences of over 120 experienced practitioners and researchers. Through this process, we were able to categorize the state of practice of data management in microservices and observe several foundational challenges that cannot be solved by software engineering practices alone, but rather require system-level support to alleviate the burden imposed on practitioners. We discuss the shortcomings of state-of-the-art database systems regarding microservices and we conclude by devising a set of features for microservice-oriented database systems.

cs.DB↗

arXiv2020

Holding a Conference Online and Live due to COVID-19

The joint EDBT/ICDT conference (International Conference on Extending Database Technology/International Conference on Database Theory) is a well established conference series on data management, with annual meetings in the second half of March that attract 250 to 300 delegates. Three weeks before EDBT/ICDT 2020 was planned to take place in Copenhagen, the rapidly developing Covid-19 pandemic led to the decision to cancel the face-to-face event. In the interest of the research community, it was decided to move the conference online while trying to preserve as much of the real-life experience as possible. As far as we know, we are one of the first conferences that moved to a fully synchronous online experience due to the COVID-19 outbreak. With fully synchronous, we mean that participants jointly listened to presentations, had live Q&A, and attended other live events associated with the conference. In this report, we share our decisions, experiences, and lessons learned.

cs.DB↗