Large-scale computing systems are increasingly using accelerators such as GPUs to enable peta-and exa-scale levels of compute to meet the needs of Machine Learning (ML) and scientific computing applications. Given the widespread and growing use of ML, including in some scientific applications, optimizing these clusters for ML workloads is particularly important. However, recent work has demonstrated that accelerators in these clusters can suffer from performance variability and this variability can lead to resource under-utilization and load imbalance. In this work we focus on how clusters schedulers, which are used to share accelerator-rich clusters across many concurrent ML jobs, can embrace performance variability to mitigate its effects. Our key insight to address this challenge is to characterize which applications are more likely to suffer from performance variability and take that into account while placing jobs on the cluster. We design a novel cluster scheduler, PAL, which uses performance variability measurements and application-specific profiles to improve job performance and resource utilization. PAL also balances performance variability with locality to ensure jobs are spread across as few nodes as possible. Overall, PAL significantly improves GPU-rich cluster scheduling: across traces for six ML workload applications spanning image, language, and vision models with a variety of variability profiles, PAL improves geomean job completion time by $42 \%$, cluster utilization by $28 \%$, and makespan by $47 \%$ over existing state-of-the-art schedulers.

The recent breakthroughs in large language models (LLMs) are positioned to transition many areas of software. In this paper, we present DB-GPT, a revolutionary and product-ready Python library that integrates LLMs into traditional data interaction tasks to enhance user experience and accessibility. DB-GPT is designed to understand data interaction tasks described by natural language and provide context-aware responses powered by LLMs, making it an indispensable tool for users ranging from novice to expert. Its system design supports deployment across local, distributed, and cloud environments. Beyond handling basic data interaction tasks like Text-to-SQL with LLMs, it can handle complex tasks like generative data analysis through a Multi-Agents framework and the Agentic Workflow Expression Language (AWEL). The Service-oriented Multi-model Management Framework (SMMF) ensures data privacy and security, enabling users to employ DB-GPT with private LLMs. Additionally, DB-GPT offers a series of product-ready features designed to enable users to integrate DB-GPT within their product environments easily. The code of DB-GPT is available at Github.

Broadening participation in computer science (CS) for primary/elementary students is a growing movement, spurred by computing workforce demands and the need for younger students to develop skills in problem solving and critical/computational thinking. However, offering computer science instruction at this level is directly related to the availability of teachers prepared to teach the subject. Unfortunately, there are relatively few primary/elementary school teachers who have received formal training in computer science, and they often self-report a lack of CS subject matter expertise. Teacher development is a key factor to address these issues, and this paper describes professional development strategies and empirical impacts of a summer institute that included two graduate courses and a series of Saturday workshops during the subsequent academic year. Key elements included teaching a high-level programing language (Python and JavaScript), integrating CS content and pedagogy instruction, and involving both experienced K-12 CS teachers and University faculty as instructors. Empirical results showed that this carefully structured PD that incorporated evidence-based elements of sufficient duration, teacher active learning and collaboration, modeling, practice, and feedback can successfully impact teacher outcomes. Results showed significant gains in teacher CS knowledge (both pedagogy and content), self-efficacy, and perception of CS value. Moderating results -- examining possible differential effects depending on teacher gender, years of teaching CS, and geographic locale -- showed that the PD was successful with experienced and less experienced teachers, with teachers from both rural and urban locales, and with both males and females.