To understand the complex interactions in modern software, engineers often rely on high-frequency telemetry (HFT) data generated via tools like eBPF. However, today's database systems are too slow for HFT's rate and volume and cannot process HFT within the limited resources available on individual host machines. Mach is a new storage engine for collecting and querying HFT. Key to Mach is the Temporal Skip Log (TSL)---a lightweight, write-optimized, log-based data structure specialized for HFT. Mach supports high ingest rates and makes data immediately queryable while operating within a limited on-host resource envelope. Our demo shows how Mach helps engineers collect and query HFT in near real-time when diagnosing performance problems. In contrast, current systems and data reduction techniques fail to keep up. While a widely used time series database (InfluxDB) drops much of the HFT, the audience will see how Mach loses no data and allows them to interactively explore HFT from application and kernel events as they arrive.

Steered query optimizers address the planning mistakes of traditional query optimizers by providing them with hints on a per-query basis, thereby guiding them in the right direction. This paper introduces QO-Insight, a visual tool designed for exploring query execution traces of such steered query optimizers. Although steered query optimizers are typically perceived as black boxes, QO-Insight empowers database administrators and experts to gain qualitative insights and enhance their performance through visual inspection and analysis.

This paper presents AutoSteer, a learning-based solution that automatically drives query optimization in any SQL database that exposes tunable optimizer knobs. AutoSteer builds on the Bandit optimizer (Bao) and extends it with new capabilities (e.g., automated hint-set discovery) to minimize integration effort and facilitate usability in both monolithic and disaggregated SQL systems. We successfully applied AutoSteer on PostgreSQL, PrestoDB, Spark-SQL, MySQL, and DuckDB - five popular open-source database engines with diverse query optimizers. We then conducted a detailed experimental evaluation with public benchmarks (JOB, Stackoverflow, TPC-DS) and a production workload from Meta's PrestoDB deployments. Our evaluation shows that AutoSteer can not only outperform these engines' native query optimizers (e.g., up to 40% improvements for PrestoDB) but can also match the performance of Bao-for-PostgreSQL with reduced human supervision and increased adaptivity, as it replaces Bao's static, expert-picked hint-sets with those that are automatically discovered. We also provide an open-source implementation of AutoSteer together with a visual tool for interactive use by query optimization experts.

Social media, self-driving cars, and traffic cameras produce video streams at large scales and cheap cost. However, storing and querying video at such scales is prohibitively expensive. We propose to treat large-scale video analytics as a data warehousing problem: Video is a format that is easy to produce but needs to be transformed into an application-specific format that is easy to query. Analogously, we define the problem of Video Extract-Transform-Load ( V-ETL ). V-ETL systems need to reduce the cost of running a user-defined V-ETL job while also giving throughput guarantees to keep up with the rate at which data is produced. We find that no current system sufficiently fulfills both needs and therefore propose Skyscraper , a system tailored to V-ETL. Skyscraper can execute arbitrary video ingestion pipelines and adaptively tunes them to reduce cost at minimal or no quality degradation, e.g., by adjusting sampling rates and resolutions to the ingested content. Skyscraper can hereby be provisioned with cheap on-premises compute and uses a combination of buffering and cloud bursting to deal with peaks in workload caused by expensive processing configurations. In our experiments, we find that Skyscraper significantly reduces the cost of V-ETL ingestion compared to adaptions of current SOTA systems, while at the same time giving robustness guarantees that these systems are lacking.

Query driven cardinality estimation models learn from a historical log of queries. They are lightweight, having low storage requirements, fast inference and training, and are easily adaptable for any kind of query. Unfortunately, such models can suffer unpredictably bad performance under workload drift, i.e., if the query pattern or data changes. This makes them unreliable and hard to deploy. We analyze the reasons why models become unpredictable due to workload drift, and introduce modifications to the query representation and neural network training techniques to make query-driven models robust to the effects of workload drift. First, we emulate workload drift in queries involving some unseen tables or columns by randomly masking out some table or column features during training. This forces the model to make predictions with missing query information, relying more on robust features based on up-to-date DBMS statistics that are useful even when query or data drift happens. Second, we introduce join bitmaps, which extends sampling-based features to be consistent across joins using ideas from sideways information passing. Finally, we show how both of these ideas can be adapted to handle data updates. We show significantly greater generalization than past works across different workloads and databases. For instance, a model trained with our techniques on a simple workload (JOBLight-train), with 40 k synthetically generated queries of at most 3 tables each, is able to generalize to the much more complex Join Order Benchmark, which include queries with up to 16 tables, and improve query runtimes by 2× over PostgreSQL. We show similar robustness results with data updates, and across other workloads. We discuss the situations where we expect, and see, improvements, as well as more challenging workload drift scenarios where these techniques do not improve much over PostgreSQL. However, even in the most challenging scenarios, our models never perform worse than PostgreSQL, while standard query driven models can get much worse than PostgreSQL.

Machine programming is an emerging research area that improves the software development life cycle from design through deployment. We present a tutorial on machine programming research highlighting aspects relevant to the data systems community. We divide this tutorial into three parts: We begin with an introduction to machine programming introducing its three pillars: intention, invention, and adaptation. Then, we provide an overview of the data ecosystem central to all machine programming systems, highlighting challenges and novel opportunities relevant to the data systems community. Finally, we describe recent advances in machine programming research and how these directions use various data sets to improve the ease of creating and maintaining performant software systems.

Recent efforts applying machine learning techniques to query optimization have shown few practical gains due to substantive training overhead, inability to adapt to changes, and poor tail performance. Motivated by these difficulties, we introduce Bao (the \underlineBa ndit \underlineo ptimizer). Bao takes advantage of the wisdom built into existing query optimizers by providing per-query optimization hints. Bao combines modern tree convolutional neural networks with Thompson sampling, a well-studied reinforcement learning algorithm. As a result, Bao automatically learns from its mistakes and adapts to changes in query workloads, data, and schema. Experimentally, we demonstrate that Bao can quickly learn strategies that improve end-to-end query execution performance, including tail latency, for several workloads containing long-running queries. In cloud environments, we show that Bao can offer both reduced costs and better performance compared with a commercial system.

Access to high-quality data repositories and benchmarks have been instrumental in advancing the state of the art in many experimental research domains. While advanced analytics tasks over time series data have been gaining lots of attention, lack of such community resources severely limits scientific progress. In this paper, we present Exathlon, the first comprehensive public benchmark for explainable anomaly detection over high-dimensional time series data. Exathlon has been systematically constructed based on real data traces from repeated executions of large-scale stream processing jobs on an Apache Spark cluster. Some of these executions were intentionally disturbed by introducing instances of six different types of anomalous events (e.g., misbehaving inputs, resource contention, process failures). For each of the anomaly instances, ground truth labels for the root cause interval as well as those for the extended effect interval are provided, supporting the development and evaluation of a wide range of anomaly detection (AD) and explanation discovery (ED) tasks. We demonstrate the practical utility of Exathlon's dataset, evaluation methodology, and end-to-end data science pipeline design through an experimental study with three state-of-the-art AD and ED techniques.

In this demo, we introduce Exathlon - a new benchmarking platform for explainable anomaly detection over high-dimensional time series. We designed Exathlon to support data scientists and researchers in developing and evaluating learned models and algorithms for detecting anomalous patterns as well as discovering their explanations. This demo will showcase Exathlon's curated anomaly dataset, novel benchmarking methodology, and end-to-end data science pipeline in action via example usage scenarios.

Recently there has been significant interest in using machine learning to improve the accuracy of cardinality estimation. This work has focused on improving average estimation error, but not all estimates matter equally for downstream tasks like query optimization. Since learned models inevitably make mistakes, the goal should be to improve the estimates that make the biggest difference to an optimizer. We introduce a new loss function, Flow-Loss, for learning cardinality estimation models. Flow-Loss approximates the optimizer's cost model and search algorithm with analytical functions, which it uses to optimize explicitly for better query plans. At the heart of Flow-Loss is a reduction of query optimization to a flow routing problem on a certain "plan graph", in which different paths correspond to different query plans. To evaluate our approach, we introduce the Cardinality Estimation Benchmark (CEB) which contains the ground truth cardinalities for sub-plans of over 16 K queries from 21 templates with up to 15 joins. We show that across different architectures and databases, a model trained with Flow-Loss improves the plan costs and query runtimes despite having worse estimation accuracy than a model trained with Q-Error. When the test set queries closely match the training queries, models trained with both loss functions perform well. However, the Q-Error-trained model degrades significantly when evaluated on slightly different queries (e.g., similar but unseen query templates), while the Flow-Loss-trained model generalizes better to such situations, achieving 4 -- 8× better 99th percentile runtimes on unseen templates with the same model architecture and training data.

New data storage technologies such as the recently introduced Intel® Optane™ DC Persistent Memory Module (PMM) offer exciting opportunities for optimizing the query processing performance of database workloads. In particular, the unique combination of low latency, byte-addressability, persistence, and large capacity make persistent memory (PMem) an attractive alternative along with DRAM and SSDs. Exploring the performance characteristics of this new medium is the first critical step in understanding how it will impact the design and performance of database systems. In this paper, we present one of the first experimental studies on characterizing Intel® Optane™ DC PMM's performance behavior in the context of analytical database workloads. First, we analyze basic access patterns common in such workloads, such as sequential, selective, and random reads as well as the complete Star Schema Benchmark, comparing standalone DRAM- and PMem-based implementations. Then we extend our analysis to join algorithms over larger datasets, which require using DRAM and PMem in a hybrid fashion while paying special attention to the read-write asymmetry of PMem. Our study reveals interesting performance tradeoffs that can help guide the design of next-generation OLAP systems in presence of persistent memory in the storage hierarchy.

Data pipelines are the new code. Consequently, data scientists need new tools to support the often time-consuming process of debugging their pipelines. We introduce Dagger , an end-to-end system to debug and mitigate data-centric errors in data pipelines, such as a data transformation gone wrong or a classifier underperforming due to noisy training data. Dagger supports inter-module debugging, where the pipeline blocks are treated as black boxes, as well as intra-module debugging, where users can debug data objects in Python scripts (e.g., DataFrames). In this demo, we will walk the audience through a rich, real-world business intelligence use case from our industrial collaborators at Intel, to highlight how Dagger enables data scientists to productively identify and mitigate data-centric problems at different stages of pipeline development.

This demo presents a novel data visualization solution for exploring the results of time series anomaly detection systems. When anomalies are reported, there is a need to reason about the results. We introduce Metro-Viz -- a visual tool to assist data scientists in performing this analysis. Metro-Viz offers a rich set of interaction features (e.g., comparative analysis, what-if testing) backed by data management strategies specifically tailored to the workload. We show our tool in action via multiple time series datasets and anomaly detectors.

Query optimization is one of the most challenging problems in database systems. Despite the progress made over the past decades, query optimizers remain extremely complex components that require a great deal of hand-tuning for specific workloads and datasets. Motivated by this shortcoming and inspired by recent advances in applying machine learning to data management challenges, we introduce Neo ( Neural Optimizer ), a novel learning-based query optimizer that relies on deep neural networks to generate query executions plans. Neo bootstraps its query optimization model from existing optimizers and continues to learn from incoming queries, building upon its successes and learning from its failures. Furthermore, Neo naturally adapts to underlying data patterns and is robust to estimation errors. Experimental results demonstrate that Neo, even when bootstrapped from a simple optimizer like PostgreSQL, can learn a model that offers similar performance to state-of-the-art commercial optimizers, and in some cases even surpass them.

This paper presents BigDAWG, a reference implementation of a new architecture for "Big Data" applications. Such applications not only call for large-scale analytics, but also for real-time streaming support, smaller analytics at interactive speeds, data visualization, and cross-storage-system queries. Guided by the principle that "one size does not fit all", we build on top of a variety of storage engines, each designed for a specialized use case. To illustrate the promise of this approach, we demonstrate its effectiveness on a hospital application using data from an intensive care unit (ICU). This complex application serves the needs of doctors and researchers and provides real-time support for streams of patient data. It showcases novel approaches for querying across multiple storage engines, data visualization, and scalable real-time analytics.

Stream processing addresses the needs of real-time applications. Transaction processing addresses the coordination and safety of short atomic computations. Heretofore, these two modes of operation existed in separate, stove-piped systems. In this work, we attempt to fuse the two computational paradigms in a single system called S-Store. In this way, S-Store can simultaneously accommodate OLTP and streaming applications. We present a simple transaction model for streams that integrates seamlessly with a traditional OLTP system, and provides both ACID and stream-oriented guarantees. We chose to build S-Store as an extension of H-Store - an open-source, in-memory, distributed OLTP database system. By implementing S-Store in this way, we can make use of the transaction processing facilities that H-Store already provides, and we can concentrate on the additional features that are needed to support streaming. Similar implementations could be done using other main-memory OLTP platforms. We show that we can actually achieve higher throughput for streaming workloads in S-Store than an equivalent deployment in H-Store alone. We also show how this can be achieved within H-Store with the addition of a modest amount of new functionality. Furthermore, we compare S-Store to two state-of-the-art streaming systems, Esper and Apache Storm, and show how S-Store can sometimes exceed their performance while at the same time providing stronger correctness guarantees.

No abstract available.

First-generation streaming systems did not pay much attention to state management via ACID transactions (e.g., [3, 4]). S-Store is a data management system that combines OLTP transactions with stream processing. To create S-Store, we begin with H-Store, a main-memory transaction processing engine, and add primitives to support streaming. This includes triggers and transaction workflows to implement push-based processing, windows to provide a way to bound the computation, and tables with hidden state to implement scoping for proper isolation. This demo explores the benefits of this approach by showing how a naïve implementation of our benchmarks using only H-Store can yield incorrect results. We also show that by exploiting push-based semantics and our implementation of triggers, we can achieve significant improvement in transaction throughput. We demo two modern applications: (i) leaderboard maintenance for a version of "American Idol", and (ii) a city-scale bicycle rental scenario.

Many stream processing applications require access to a multitude of streaming as well as stored data sources. Yet there is no clear semantics for correct continuous query execution over these data sources in the face of concurrent access and failures. Instead, today's Stream Processing Systems (SPSs) hard-code transactional concepts in their execution models, making them both hard to understand and inflexible to use. In this paper, we show that we can successfully reuse the traditional transactional theory (with some minimal extensions) in order to cleanly define the correct interaction of a set of continuous and one-time queries concurrently accessing both streaming and stored data sources. The result is a unified transactional model (UTM) for query processing over streams as well as traditional databases. We present a transaction manager that implements this model on top of an existing storage manager for streams (MXQuery/SMS). Experiments on the Linear Road Benchmark show that our transaction manager flexibly ensures correctness in case of concurrency and failures, without sacrificing from performance. Moreover, this model is powerful enough to express the implicit transactional behaviors of a representative set of state-of-the-art SPSs.

Continuous queries can run for unpredictably long periods of time. During their lifetime, these queries may need to be adapted either due to changes in application semantics (e.g., the implementation of a new alert detection policy), or due to changes in the system's behavior (e.g., adapting performance to a changing load). While in previous works query modification has been implicitly utilized to serve specific purposes (e.g., load management), to date no research has been done that defines a general-purpose, reliable, and efficiently implementable model for modifying continuous queries at run-time. In this paper, we introduce a punctuation-based framework that can formally express arbitrary lifecycle operations on the basis of input-output mappings and basic control elements such as start or stop of queries. On top of this foundation, we derive all possible query change methods, each providing different levels of correctness guarantees and performance. We further show how these models can be efficiently realized in a state-of-the-art stream processing engine; we also provide experimental results demonstrating the key performance tradeoffs of the change methods.

MaxStream is a federated stream processing system that seamlessly integrates multiple autonomous and heterogeneous Stream Processing Engines (SPEs) and databases. In this paper, we propose to demonstrate the key features of MaxStream using two application scenarios, namely the Sales Map & Spikes business monitoring scenario and the Linear Road Benchmark, each with a different set of requirements. More specifically, we will show how the MaxStream Federator can translate and forward the application queries to two different commercial SPEs (Coral8 and StreamBase), as well as how it does so under various persistency requirements.

There are many academic and commercial stream processing engines (SPEs) today, each of them with its own execution semantics. This variation may lead to seemingly inexplicable differences in query results. In this paper, we present SECRET, a model of the behavior of SPEs. SECRET is a descriptive model that allows users to analyze the behavior of systems and understand the results of window-based queries for a broad range of heterogeneous SPEs. The model is the result of extensive analysis and experimentation with several commercial and academic engines. In the paper, we describe the types of heterogeneity found in existing engines, and show with experiments on real systems that our model can explain the key differences in windowing behavior.

Data Stream Management Systems (DSMS) operate under strict performance requirements. Key to meeting such requirements is to efficiently handle time-critical tasks such as managing internal states of continuous query operators, traffic on the queues between operators, as well as providing storage support for shared computation and archived data. In this paper, we introduce a general purpose storage management framework for DSMSs that performs these tasks based on a clean, loosely-coupled, and flexible system design that also facilitates performance optimization. An important contribution of the framework is that, in analogy to buffer management techniques in relational database systems, it uses information about the access patterns of streaming applications to tune and customize the performance of the storage manager. In the paper, we first analyze typical application requirements at different granularities in order to identify important tunable parameters and their corresponding values. Based on these parameters, we define a general-purpose storage management interface. Using the interface, a developer can use our SMS (Storage Manager for Streams) to generate a customized storage manager for streaming applications. We explore the performance and potential of SMS through a set of experiments using the Linear Road benchmark.

DejaVu is an event processing system that integrates declarative pattern matching over live and archived streams of events on top of a novel system architecture. We propose to demonstrate the key aspects of the DejaVu query language and architecture using two different application scenarios, namely a smart RFID library system and a financial market data analysis application. The demonstration will illustrate how DejaVu can uniformly handle one-time, continuous, and hybrid pattern matching queries over live and archived stream stores, using highly interactive visual monitoring tools including one that is based on the Second Life virtual world.

Complex Event Detection (CED) is emerging as a key capability for many monitoring applications such as intrusion detection, sensor-based activity & phenomena tracking, and network monitoring. Existing CED solutions commonly assume centralized availability and processing of all relevant events, and thus incur significant overhead in distributed settings. In this paper, we present and evaluate communication efficient techniques that can efficiently perform CED across distributed event sources. Our techniques are plan-based: we generate multi-step event acquisition and processing plans that leverage temporal relationships among events and event occurrence statistics to minimize event transmission costs, while meeting application-specific latency expectations. We present an optimal but exponential-time dynamic programming algorithm and two polynomial-time heuristic algorithms, as well as their extensions for detecting multiple complex events with common sub-expressions. We characterize the behavior and performance of our solutions via extensive experimentation on synthetic and real-world data sets using our prototype implementation.

Borealis is a distributed stream processing engine that is being developed at Brandeis University, Brown University, and MIT. Borealis inherits core stream processing functionality from Aurora and inter-node communication functionality from Medusa.We propose to demonstrate some of the key aspects of distributed operation in Borealis, using a multi-player network game as the underlying application. The demonstration will illustrate the dynamic resource management, query optimization and high availability mechanisms employed by Borealis, using visual performance-monitoring tools as well as the gaming experience.

No abstract available.

MARIFlow System [1] provides for automating and monitoring the flow of control and data over the Internet among different organizations, thereby creating a platform necessary to describe higher order processes involving several organizations and companies. The architecture is general enough to be applied to any business practice where data flow among different industries and cooperations and the invocation of activities follow a pattern that can be described through a process definition. The example application provided within the scope of this project is on maritime industry A MARIFlow process is executed through cooperating agents, called MARCAs (MARIFlow Cooperating Agents) that are automatically initialized at each site that the process executes. MARCAs handle the activities at their site, provide for coordination with other MARCAs in the system by routing the documents in electronic form according to the process description, keeping track of process information, and providing for the security and authentication of documents as well as comprehensive monitoring facilities. More specifically, the functionality provided by the system is as follows: ¯ A declarative means to specify the control of document flow over the Internet where it is possible to define the source of data, its control flow and the activities that make use of this data. ¯ A fully distributed execution architecture achieved through cooperating agents over the Internet. The agents know about other agents that they need to communicate with and preserve their state during communication. They also manage local information for monitoring purposes and for recovering from failures. ¯ Communicating with inside firewall applications. A MARCA can activate in-house activities automati-