Is the GPU Half-Empty or Half-Full? Practical Scheduling Techniques for LLMs

Serving systems for Large Language Models (LLMs) improve throughput by processing several requests concurrently. However, multiplexing hardware resources between concurrent requests involves non-trivial scheduling decisions. Practical serving systems typically implement these decisions at two levels: First, a load balancer routes requests to different servers which each hold a replica of the LLM. Then, on each server, an engine-level scheduler decides when to run a request, or when to queue or preempt it. Improved scheduling policies may benefit a wide range of LLM deployments and can often be implemented as "drop-in replacements" to a system's current policy. In this work, we survey scheduling techniques from the literature and from practical serving systems. We find that schedulers from the literature often achieve good performance but introduce significant complexity. In contrast, schedulers in practical deployments often leave easy performance gains on the table but are easy to implement, deploy and configure. This finding motivates us to introduce two new scheduling techniques, which are both easy to implement, and outperform current techniques on production workload traces.

BEAVER: An Enterprise Benchmark for Text-to-SQL

Existing text-to-SQL benchmarks have largely been constructed using publicly available tables from the web with human-generated tests containing question and SQL statement pairs. They typically show very good results and lead people to think that LLMs are effective at text-to-SQL tasks. In this paper, we apply off-the-shelf LLMs to a benchmark containing enterprise data warehouse data. In this environment, LLMs perform poorly, even when standard prompt engineering and RAG techniques are utilized. As we will show, the reasons for poor performance are largely due to three characteristics: (1) public LLMs cannot train on enterprise data warehouses because they are largely in the "dark web", (2) schemas of enterprise tables are more complex than the schemas in public data, which leads the SQL-generation task innately harder, and (3) business-oriented questions are often more complex, requiring joins over multiple tables and aggregations. As a result, we propose a new dataset BEAVER, sourced from real enterprise data warehouses together with natural language queries and their correct SQL statements which we collected from actual user history. We evaluated this dataset using recent LLMs and demonstrated their poor performance on this task. We hope this dataset will facilitate future researchers building more sophisticated text-to-SQL systems which can do better on this important class of data.

MDCR: A Dataset for Multi-Document Conditional Reasoning

The same real-life questions posed to different individuals may lead to different answers based on their unique situations. For instance, whether a student is eligible for a scholarship depends on eligibility conditions, such as major or degree required. ConditionalQA was proposed to evaluate models' capability of reading a document and answering eligibility questions, considering unmentioned conditions. However, it is limited to questions on single documents, neglecting harder cases that may require cross-document reasoning and optimization, for example, "What is the maximum number of scholarships attainable?" Such questions over multiple documents are not only more challenging due to more context having to understand, but also because the model has to (1) explore all possible combinations of unmentioned conditions and (2) understand the relationship between conditions across documents, to reason about the optimal outcome. To evaluate models' capability of answering such questions, we propose a new dataset MDCR, which can reflect real-world challenges and serve as a new test bed for complex conditional reasoning that requires optimization. We evaluate this dataset using the most recent LLMs and demonstrate their limitations in solving this task. We believe this dataset will facilitate future research in answering optimization questions with unknown conditions.

A Declarative System for Optimizing AI Workloads

Modern AI models provide the key to a long-standing dream: processing analytical queries about almost any kind of data. Until recently, it was difficult and expensive to extract facts from company documents, data from scientific papers, or insights from image and video corpora. Today's models can accomplish these tasks with high accuracy. However, a programmer who wants to answer a substantive AI-powered query must orchestrate large numbers of models, prompts, and data operations. For even a single query, the programmer has to make a vast number of decisions such as the choice of model, the right inference method, the most cost-effective inference hardware, the ideal prompt design, and so on. The optimal set of decisions can change as the query changes and as the rapidly-evolving technical landscape shifts. In this paper we present Palimpzest, a system that enables anyone to process AI-powered analytical queries simply by defining them in a declarative language. The system uses its cost optimization framework -- which explores the search space of AI models, prompting techniques, and related foundation model optimizations -- to implement the query with the best trade-offs between runtime, financial cost, and output data quality. We describe the workload of AI-powered analytics tasks, the optimization methods that Palimpzest uses, and the prototype system itself. We evaluate Palimpzest on tasks in Legal Discovery, Real Estate Search, and Medical Schema Matching. We show that even our simple prototype offers a range of appealing plans, including one that is 3.3x faster, 2.9x cheaper, and offers better data quality than the baseline method. With parallelism enabled, Palimpzest can produce plans with up to a 90.3x speedup at 9.1x lower cost relative to a single-threaded GPT-4 baseline, while obtaining an F1-score within 83.5% of the baseline. These require no additional work by the user.

SEED: Simple, Efficient, and Effective Data Management via Large Language Models

We introduce SEED, an LLM-centric system that allows users to easily create efficient, and effective data management applications. SEED comprises three main components: code generation, model generation, and augmented LLM query to address the challenges that LLM services are computationally and economically expensive and do not always work well on all cases for a given data management task. SEED addresses the expense challenge by localizing LLM computation as much as possible. This includes replacing most of LLM calls with local code, local models, and augmenting LLM queries with batching and data access tools, etc. To ensure effectiveness, SEED features a bunch of optimization techniques to enhance the localized solution and the LLM queries, including automatic code validation, code ensemble, model representatives selection, selective tool usages, etc. Moreover, with SEED users are able to easily construct a data management solution customized to their applications. It allows the users to configure each component and compose an execution pipeline in natural language. SEED then automatically compiles it into an executable program. We showcase the efficiency and effectiveness of SEED using diverse data management tasks such as data imputation, NL2SQL translation, etc., achieving state-of-the-art few-shot performance while significantly reducing the number of required LLM calls.

TagMe: GPS-Assisted Automatic Object Annotation in Videos

Training high-accuracy object detection models requires large and diverse annotated datasets. However, creating these data-sets is time-consuming and expensive since it relies on human annotators. We design, implement, and evaluate TagMe, a new approach for automatic object annotation in videos that uses GPS data. When the GPS trace of an object is available, TagMe matches the object's motion from GPS trace and the pixels' motions in the video to find the pixels belonging to the object in the video and creates the bounding box annotations of the object. TagMe works using passive data collection and can continuously generate new object annotations from outdoor video streams without any human annotators. We evaluate TagMe on a dataset of 100 video clips. We show TagMe can produce high-quality object annotations in a fully-automatic and low-cost way. Compared with the traditional human-in-the-loop solution, TagMe can produce the same amount of annotations at a much lower cost, e.g., up to 110x.

Technical Report on Data Integration and Preparation

AI application developers typically begin with a dataset of interest and a vision of the end analytic or insight they wish to gain from the data at hand. Although these are two very important components of an AI workflow, one often spends the first few weeks (sometimes months) in the phase we refer to as data conditioning. This step typically includes tasks such as figuring out how to prepare data for analytics, dealing with inconsistencies in the dataset, and determining which algorithm (or set of algorithms) will be best suited for the application. Larger, faster, and messier datasets such as those from Internet of Things sensors, medical devices or autonomous vehicles only amplify these issues. These challenges, often referred to as the three Vs (volume, velocity, variety) of Big Data, require low-level tools for data management, preparation and integration. In most applications, data can come from structured and/or unstructured sources and often includes inconsistencies, formatting differences, and a lack of ground-truth labels. In this report, we highlight a number of tools that can be used to simplify data integration and preparation steps. Specifically, we focus on data integration tools and techniques, a deep dive into an exemplar data integration tool, and a deep-dive in the evolving field of knowledge graphs. Finally, we provide readers with a list of practical steps and considerations that they can use to simplify the data integration challenge. The goal of this report is to provide readers with a view of state-of-the-art as well as practical tips that can be used by data creators that make data integration more seamless.

Unnatural Language Processing: Bridging the Gap Between Synthetic and Natural Language Data

Large, human-annotated datasets are central to the development of natural language processing models. Collecting these datasets can be the most challenging part of the development process. We address this problem by introducing a general purpose technique for ``simulation-to-real'' transfer in language understanding problems with a delimited set of target behaviors, making it possible to develop models that can interpret natural utterances without natural training data. We begin with a synthetic data generation procedure, and train a model that can accurately interpret utterances produced by the data generator. To generalize to natural utterances, we automatically find projections of natural language utterances onto the support of the synthetic language, using learned sentence embeddings to define a distance metric. With only synthetic training data, our approach matches or outperforms state-of-the-art models trained on natural language data in several domains. These results suggest that simulation-to-real transfer is a practical framework for developing NLP applications, and that improved models for transfer might provide wide-ranging improvements in downstream tasks.

Physical Representation-based Predicate Optimization for a Visual Analytics Database

Querying the content of images, video, and other non-textual data sources requires expensive content extraction methods. Modern extraction techniques are based on deep convolutional neural networks (CNNs) and can classify objects within images with astounding accuracy. Unfortunately, these methods are slow: processing a single image can take about 10 milliseconds on modern GPU-based hardware. As massive video libraries become ubiquitous, running a content-based query over millions of video frames is prohibitive. One promising approach to reduce the runtime cost of queries of visual content is to use a hierarchical model, such as a cascade, where simple cases are handled by an inexpensive classifier. Prior work has sought to design cascades that optimize the computational cost of inference by, for example, using smaller CNNs. However, we observe that there are critical factors besides the inference time that dramatically impact the overall query time. Notably, by treating the physical representation of the input image as part of our query optimization---that is, by including image transforms, such as resolution scaling or color-depth reduction, within the cascade---we can optimize data handling costs and enable drastically more efficient classifier cascades. In this paper, we propose Tahoma, which generates and evaluates many potential classifier cascades that jointly optimize the CNN architecture and input data representation. Our experiments on a subset of ImageNet show that Tahoma's input transformations speed up cascades by up to 35 times. We also find up to a 98x speedup over the ResNet50 classifier with no loss in accuracy, and a 280x speedup if some accuracy is sacrificed.

Database Learning: Toward a Database that Becomes Smarter Every Time

In today's databases, previous query answers rarely benefit answering future queries. For the first time, to the best of our knowledge, we change this paradigm in an approximate query processing (AQP) context. We make the following observation: the answer to each query reveals some degree of knowledge about the answer to another query because their answers stem from the same underlying distribution that has produced the entire dataset. Exploiting and refining this knowledge should allow us to answer queries more analytically, rather than by reading enormous amounts of raw data. Also, processing more queries should continuously enhance our knowledge of the underlying distribution, and hence lead to increasingly faster response times for future queries. We call this novel idea---learning from past query answers---Database Learning. We exploit the principle of maximum entropy to produce answers, which are in expectation guaranteed to be more accurate than existing sample-based approximations. Empowered by this idea, we build a query engine on top of Spark SQL, called Verdict. We conduct extensive experiments on real-world query traces from a large customer of a major database vendor. Our results demonstrate that Verdict supports 73.7% of these queries, speeding them up by up to 23.0x for the same accuracy level compared to existing AQP systems.