GreenMix is motivated by the renewed interest in asymmetric multi-core processors and the emergence of the serverless computing model. Asymmetric multi-cores offer better energy and performance trade-offs by placing different core types on the same die. However, existing serverless scheduling techniques do not leverage these benefits. GreenMix is the first serverless work to reduce energy and serverless keep-alive costs while meeting QoS targets by leveraging asymmetric multi-cores. GreenMix employs randomized sketching, tailored for serverless execution and keep-alive, to perform within 10% of the optimal solution in terms of energy efficiency and keep-alive cost reduction. GreenMix’s effectiveness is demonstrated through evaluations on clusters of ARM big.LITTLE and Intel Alder Lake asymmetric processors. It outperforms competing state-of-the-art schedulers, offering a novel approach for energy-efficient serverless computing.

The rapid advancement of generative AI has heightened environmental concerns, particularly regarding carbon emissions.Our framework, SPROUT, addresses these challenges by reducing the carbon footprint of inference in large language models (LLMs).SPROUT introduces "generation directives" to guide the autoregressive generation process, achieving a balance between ecological sustainability and high-quality outputs.By employing a strategic optimizer for directive assignment and a novel offline quality evaluator, SPROUT reduces the carbon footprint of generative LLM inference by over 40% in real-world evaluations, using the Llama model and global electricity grid data.This work is crucial as the rising interest in inference time compute scaling laws amplifies environmental concerns, emphasizing the need for eco-friendly AI solutions.

Modern high-performance computing (HPC) and cloud computing systems are integrating powerful GPUs to accelerate increasingly demanding deep learning workloads. To improve cluster efficiency and better understand user behavior and job characteristics, system operators will collect operational data for trace analysis. However, previous efforts on these system logs have lacked the interpretability aspect, and there is no systematic approach that can be widely applied to different datacenter traces and return interpretable results. In this work, we propose a workflow to discover hidden association relation-ships between collected features of system jobs. The outcome of our analysis approach yields useful association rules that can be directly interpreted into operational insights. Using this approach, we have conducted case studies using the traces of three large-scale multi-tenant GPU clusters running production machine learning workloads. We have focused on the observations of GPU underutilization and job failures, revealing the possible reasons for these job behaviors and suggesting solutions to mitigate them. Our case studies have demonstrated the feasibility of our interpretable analysis workflow, which can be widely used by more HPC and cloud computing system operators.

As research and deployment of AI grows, the computational burden to support and sustain its progress inevitably does too. To train or fine-tune state-of-the-art models in NLP, computer vision, etc., some form of AI hardware acceleration is virtually a requirement. Recent large language models require considerable resources to train and deploy, resulting in significant energy usage, potential carbon emissions, and massive demand for GPUs and other hardware accelerators. However, this surge carries large implications for energy sustainability at the HPC/datacenter level. In this paper, we study the effects of power-capping GPUs at a research supercomputing center on GPU temperature and power draw; we show significant decreases in both temperature and power draw, reducing power consumption and potentially improving hardware life-span, with minimal impact on job performance. To our knowledge, our work is the first to conduct and make available a detailed analysis of the effects of GPU power-capping at the supercomputing scale. We hope our work will inspire HPCs/datacenters to further explore, evaluate, and communicate the impact of power-capping AI hardware accelerators for more sustainable AI.

Online inference is becoming a key service product for many businesses, deployed in cloud platforms to meet customer demands. Despite their revenue-generation capability, these services need to operate under tight Quality-of-Service (QoS) and cost budget constraints. This paper introduces KAIROS, a novel runtime framework that maximizes the query throughput while meeting QoS target and a cost budget. KAIROS designs and implements novel techniques to build a pool of heterogeneous compute hardware without online exploration overhead, and distribute inference queries optimally at runtime. Our evaluation using industry-grade machine learning (ML) models shows that KAIROS yields up to 2x the throughput of an optimal homogeneous solution, and outperforms state-of-the-art schemes by up to 70%, despite advantageous implementations of the competing schemes to ignore their exploration overhead.

The rapid growth in demand for HPC systems has led to a rise in carbon footprint, which requires urgent intervention. In this work, we present a comprehensive analysis of the carbon footprint of highperformance computing (HPC) systems, considering the carbon footprint during both the hardware manufacturing and system operational stages. Our work employs HPC hardware component carbon footprint modeling, regional carbon intensity analysis, and experimental characterization of the system life cycle to highlight the importance of quantifying the carbon footprint of HPC systems.

This paper presents a solution to the challenge of mitigating carbon emissions from hosting large-scale machine learning (ML) inference services. ML inference is critical to modern technology products, but it is also a significant contributor to carbon footprint. We introduce, Clover, a carbon-friendly ML inference service runtime system that balances performance, accuracy, and carbon emissions through mixed-quality models and GPU resource partitioning. Our experimental results demonstrate that Clover is effective in substantially reducing carbon emissions while maintaining high accuracy and meeting service level agreement (SLA) targets.

As image datasets become ubiquitous, the problem of ad-hoc searches over image data is increasingly important. Many high-level data tasks in machine learning, such as constructing datasets for training and testing object detectors, imply finding ad-hoc objects or scenes within large image datasets as a key sub-problem. New foundational visual-semantic embeddings trained on massive web datasets such as Contrastive Language-Image Pre-Training (CLIP) can help users start searches on their own data, but we find there is a long tail of queries where these models fall short in practice. Seesaw is a system for interactive ad-hoc searches on image datasets that integrates state-of-the-art embeddings like CLIP with user feedback in the form of box annotations to help users quickly locate images of interest in their data even in the long tail of harder queries. One key challenge for Seesaw is that, in practice, many sensible approaches to incorporating feedback into future results, including state-of-the-art active-learning algorithms, can worsen results compared to introducing no feedback, partly due to CLIP's high-average performance. Therefore, Seesaw includes several algorithms that empirically result in larger and also more consistent improvements. We compare Seesaw's accuracy to both using CLIP alone and to a state-of-the-art active-learning baseline and find Seesaw consistently helps improve results for users across four datasets and more than a thousand queries. Seesaw increases Average Precision (AP) on search tasks by an average of .08 on a wide benchmark (from a base of .72), and by a .27 on a subset of more difficult queries where CLIP alone performs poorly.

GPU technology has been improving at an expedited pace in terms of size and performance, empowering HPC and AI/ML researchers to advance the scientific discovery process. However, this also leads to inefficient resource usage, as most GPU workloads, including complicated AI/ML models, are not able to utilize the GPU resources to their fullest extent - encouraging support for GPU multi-tenancy. We propose MISO, a technique to exploit the Multi-Instance GPU (MIG) capability on the latest NVIDIA datacenter GPUs (e.g., A100, H100) to dynamically partition GPU resources among co-located jobs. MISO's key insight is to use the lightweight, more flexible Multi-Process Service (MPS) capability to predict the best MIG partition allocation for different jobs, without incurring the overhead of implementing them during exploration. Due to its ability to utilize GPU resources more efficiently, MISO achieves 49% and 16% lower average job completion time than the unpartitioned and optimal static GPU partition schemes, respectively.

Production high-performance computing (HPC) systems are adopting and integrating GPUs into their design to accommodate artificial intelligence (AI), machine learning, and data visualization workloads. To aid with the design and operations of new and existing GPU-based large-scale systems, we provide a detailed characterization of system operations, job characteristics, user behavior, and trends on a contemporary GPU-accelerated production HPC system. Our insights indicate that the pre-mature phases in modern AI workflow take up significant GPU hours while underutilizing GPUs, which opens up the opportunity for a multi-tier system. Finally, we provide various potential recommendations and areas for future investment for system architects, operators, and users.

Capturing and processing video is increasingly common as cameras become cheaper to deploy. At the same time, rich video-understanding methods have progressed greatly in the last decade. As a result, many organizations now have massive repositories of video data, with applications in mapping, navigation, autonomous driving, and other areas. Because state-of-the-art object-detection methods are slow and expensive, our ability to process even simple ad-hoc object search queries (“find 100 traffic lights in dashcam video”) over this accumulated data lags far behind our ability to collect the data. Processing video at reduced sampling rates is a reasonable default strategy for these types of queries; however, the ideal sampling rate is both data and query dependent. We introduce ExSample, a low cost framework for object search over un-indexed video that quickly processes search queries by adapting the amount and location of sampled frames to the particular data and query being processed. ExSample prioritizes the processing of frames in a video repository so that processing is focused in portions of video that most likely contain objects of interest. It approaches searching in a similar way to a multi-arm bandit problem where each arm corresponds to a portion of a video. On large, real-world datasets, ExSample reduces processing time by 1.9x on average and up to 6x over an efficient random sampling baseline. Moreover, we show ExSample finds many results long before sophisticated, state-of-the-art baselines based on proxy scores can begin producing their first results.

Capturing and processing video is increasingly com-mon as cameras become cheaper to deploy. At the same time, rich video-understanding methods have progressed greatly in the last decade. As a result, many organizations now have massive repositories of video data, with applications in mapping, navigation, autonomous driving, and other areas. Because state-of-the-art object-detection methods are slow and expensive, our ability to process even simple ad-hoc object search queries (“find 100 traffic lights in dashcam video”) over this accumulated data lags far behind our ability to collect the data. Processing video at reduced sampling rates is a reasonable default strategy for these types of queries; however, the ideal sampling rate is both data and query dependent. We introduce ExSample, a low cost framework for object search over un-indexed video that quickly processes search queries by adapting the amount and location of sampled frames to the particular data and query being processed. ExSample prioritizes the processing of frames in a video repository so that processing is focused in portions of video that most likely contain objects of interest. It approaches searching in a similar way to a multi-arm bandit problem where each arm corresponds to a portion of a video. On large, real-world datasets, ExSample reduces processing time by 1.9x on average and up to 6x over an efficient random sampling baseline. Moreover, we show ExSample finds many results long before sophisticated, state-of-the-art baselines based on proxy scores can begin producing their first results.

This work introduces Mashup, a novel strategy to leverage serverless computing model for executing scientific workflows in a hybrid fashion by taking advantage of both the traditional VM-based cloud computing platform and the emerging serverless platform. Mashup outperforms the state-of-the-art workflow execution engines by an average of 34% and 43% in terms of execution time reduction and cost reduction, respectively, for widely-used HPC workflows on the Amazon Cloud platform (EC2 and Lambda).

As parallel applications become more complex, auto-tuning becomes more desirable, challenging, and time-consuming. We propose, Bliss, a novel solution for auto-tuning parallel applications without requiring apriori information about applications, domain-specific knowledge, or instrumentation. Bliss demonstrates how to leverage a pool of Bayesian Optimization models to find the near-optimal parameter setting 1.64× faster than the state-of-the-art approaches.

Deep learning model inference is a key service in many businesses and scientific discovery processes. This paper introduces Ribbon, a novel deep learning inference serving system that meets two competing objectives: quality-of-service (QoS) target and cost-effectiveness. The key idea behind Ribbon is to intelligently employ a diverse set of cloud computing instances (heterogeneous instances) to meet the QoS target and maximize cost savings. Ribbon devises a Bayesian Optimization-driven strategy that helps users build the optimal set of heterogeneous instances for their model inference service needs on cloud computing platforms - and, Ribbon demonstrates its superiority over existing approaches of inference serving systems using homogeneous instance pools. Ribbon saves up to 16% of the inference service cost for different learning models including emerging deep learning recommender system models and drug-discovery enabling models.

In order to conduct analytical tasks, data scientists often need to find relevant data from an avalanche of sources (e.g., data lakes, large organizational databases). This effort is typically made in an ad hoc, non-systematic manner, which makes it a daunting endeavour. Current data discovery systems typically require the users to find relevant tables manually, usually by issuing multiple queries (e.g., using SQL). However, expressing such queries is nontrivial, as it requires knowledge of the underlying structure (schema) of the data organization in advance. This issue is further exacerbated when data resides in data lakes, where there is no predefined schema that data must conform to. On the other hand, data scientists can often come up with a few example records of interest quickly. Motivated by this observation, we developed DICE---a human-in-the-loop system for Data dIsCovery by Example---that takes user-provided example records as input and returns more records that satisfy the user intent. DICE's key idea is to synthesize a SQL query that captures the user intent, specified via examples. To this end, DICE follows a three-step process: (1) DICE first discovers a few candidate queries by finding join paths across tables within the data lake. (2) Then DICE consults with the user for validation by presenting a few records to them, and, thus, eliminating spurious queries. (3) Based on the user feedback, DICE refines the search and repeats the process until the user is satisfied with the results. We will demonstrate how DICE can help in data discovery through an interactive, example-based interaction.

This paper lays out the rationale for building a completely new operating system (OS) stack. Rather than build on a single node OS together with separate cluster schedulers, distributed filesystems, and network managers, we argue that a distributed transactional DBMS should be the basis for a scalable cluster OS. We show herein that such a database OS (DBOS) can do scheduling, file management, and inter-process communication with competitive performance to existing systems. In addition, significantly better analytics can be provided as well as a dramatic reduction in code complexity through implementing OS services as standard database queries, while implementing low-latency transactions and high availability only once.

Detecting and classifying anomalous behaviors in computer networks remains a formidable challenge. This work outlines a machine learning technique that uses deep neural networks to detect and classify a variety of network attacks. Our approach is based on that hypothesis that different network attacks generate a distinguishable change in entropy of certain network flow features. To generate a training and validation dataset, we inject synthetic attacks of different types and intensities into raw packet capture data collected from an internet backbone link by the MAWI group. Experimental results show that our machine learning classification model can achieve high accuracy for network attacks in which attack intensities are as low as 5% of overall traffic.

Financial transactions, internet search, and data analysis are all placing increasing demands on databases. SQL, NoSQL, and NewSQL databases have been developed to meet these demands and each offers unique benefits. SQL, NoSQL, and NewSQL databases also rely on different underlying mathematical models. Polystores seek to provide a mechanism to allow applications to transparently achieve the benefits of diverse databases while insulating applications from the details of these databases. Integrating the underlying mathematics of these diverse databases can be an important enabler for polystores as it enables effective reasoning across different databases. Associative arrays provide a common approach for the mathematics of polystores by encompassing the mathematics found in different databases: sets (SQL), graphs (NoSQL), and matrices (NewSQL). Prior work presented the SQL relational model in terms of associative arrays and identified key mathematical properties that are preserved within SQL. This work provides the rigorous mathematical definitions, lemmas, and theorems underlying these properties. Specifically, SQL Relational Algebra deals primarily with relations - multisets of tuples - and operations on and between those relations. These relations can be modeled as associative arrays by treating tuples as non-zero rows in an array. Operations in relational algebra are built as compositions of standard operations on associative arrays which mirror their matrix counterparts. These constructions provide insight into how relational algebra can be handled via array operations. As an example application, the composition of two projection operations is shown to also be a projection, and the projection of a union is shown to be equal to the union of the projections.

Modern applications often need to manage and analyze widely diverse datasets that span multiple data models [1], [2], [3], [4], [5]. Warehousing the data through Extract-Transform-Load (ETL) processes can be expensive in such scenarios. Transforming disparate data into a single data model may degrade performance. Further, curating diverse datasets and maintaining the pipeline can prove to be labor intensive. As a result, an emerging trend is to shift the focus to federating specialized data stores and enabling query processing across heterogeneous data models [6]. This shift can bring many advantages: First, systems can natively leverage multiple data models, which can translate to maximizing the semantic expressiveness of underlying interfaces and leveraging the internal processing capabilities of component data stores. Second, federated architectures support query-specific data integration with just-in-time transformation and migration, which has the potential to significantly reduce the operational complexity and overhead. Projects that focus on developing systems in this research area stem from various backgrounds and address diverse concerns, which could make it difficult to form a consistent view of the work in this area. In this survey, we introduce a taxonomy for describing the state of the art and propose a systematic evaluation framework conducive to understanding of query-processing characteristics in the relevant systems. We use the framework to assess four representative implementations: BigDAWG [7], [8], CloudMdsQL [9], [10], Myria [11], [12], and Apache Drill [13].

Protected database search systems cryptographically isolate the roles of reading from, writing to, and administering the database. This separation limits unnecessary administrator access and protects data in the case of system breaches. Since protected search was introduced in 2000, the area has grown rapidly, systems are offered by academia, start-ups, and established companies. However, there is no best protected search system or set of techniques. Design of such systems is a balancing act between security, functionality, performance, and usability. This challenge is made more difficult by ongoing database specialization, as some users will want the functionality of SQL, NoSQL, or NewSQL databases. This database evolution will continue, and the protected search community should be able to quickly provide functionality consistent with newly invented databases. At the same time, the community must accurately and clearly characterize the tradeoffs between different approaches. To address these challenges, we provide the following contributions:1) An identification of the important primitive operations across database paradigms. We find there are a small number of base operations that can be used and combined to support a large number of database paradigms.2) An evaluation of the current state of protected search systems in implementing these base operations. This evaluation describes the main approaches and tradeoffs for each base operation. Furthermore, it puts protected search in the context of unprotected search, identifying key gaps in functionality.3) An analysis of attacks against protected search for different base queries.4) A roadmap and tools for transforming a protected search system into a protected database, including an open-source performance evaluation platform and initial user opinions of protected search.

State of the art sensors within a single autonomous vehicle (AV) can produce video and LIDAR data at rates greater than 30 GB/hour. Unsurprisingly, even small AV research teams can accumulate tens of terabytes of sensor data from multiple trips and multiple vehicles. AV practitioners would like to extract information about specific locations or specific situations for further study, but are often unable to. Queries over AV sensor data are different from generic analytics or spatial queries because they demand reasoning about fields of view as well as heavy computation to extract features from scenes. In this article and demo we present Vroom, a system for ad-hoc queries over AV sensor databases. Vroom combines domain specific properties of AV datasets with selective indexing and multi-query optimization to address challenges posed by AV sensor data.

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.

We present a framework for the estimation of driver behavior at intersections, with applications to autonomous driving and vehicle safety. The framework is based on modeling the driver behavior and vehicle dynamics as a hybrid-state system (HSS), with driver decisions being modeled as a discrete-state system and the vehicle dynamics modeled as a continuous-state system. The proposed estimation method uses observable parameters to track the instantaneous continuous state and estimates the most likely behavior of a driver given these observations. This paper describes a framework that encompasses the hybrid structure of vehicle-driver coupling and uses hidden Markov models (HMMs) to estimate driver behavior from filtered continuous observations. Such a method is suitable for scenarios that involve unknown decisions of other vehicles, such as lane changes or intersection access. Such a framework requires extensive data collection, and the authors describe the procedure used in collecting and analyzing vehicle driving data. For illustration, the proposed hybrid architecture and driver behavior estimation techniques are trained and tested near intersections with exemplary results provided. Comparison is made between the proposed framework, simple classifiers, and naturalistic driver estimation. Obtained results show promise for using the HSS-HMM framework.