In this paper, we describe veDB-HTAP, a highly integrated, efficient, and adaptive HTAP system recently built in ByteDance. veDB-HTAP adopts a highly integrated system architecture by leveraging the Secondary Engine mechanism provided by MySQL and provides a seamless query processing experience across OLTP and OLAP engines. In addition, we introduce a cost-based and machine-learning-based smart query router that significantly outperforms the rule-based query router used in ByteHTAP, a precursor of veDB-HTAP. A key design principle of veDB-HTAP is the collaboration and adaptability of major system components, including query planning, query execution, and unified storage. Our adaptive query execution can be classified into two categories: 1) adaptive execution that dynamically collects and utilizes runtime statistics for better query performance; 2) utilizing runtime resource information to achieve a high quality of service even under heavy workloads. The experiments show that veDB-HTAP can achieve more than 3× speedup for TPC-H while consuming only one-third of the resources compared to ByteHTAP.

Machine learning (ML) has become a cornerstone of key applications at ByteDance. As model complexity and data volumes surge, data management for large-scale ML workloads faces substantial challenges, particularly with recent advances in large recommendation models (LRMs) and large multimodal models (LMMs). Traditional approaches exhibit limitations in storage efficiency, metadata scalability, update mechanisms, and integration with ML frameworks. To address these challenges, we propose Magnus, a holistic data management system built upon Apache Iceberg. Magnus integrates innovative optimizations across resource-efficient storage formats optimized for large wide tables and multimodal data, built-in support for vector and inverted indexes to accelerate data retrieval, scalable metadata planning with Git-like branching and tagging capabilities, and high-performance update/upsert based on lightweight merge-on-read (MOR) strategies. Additionally, Magnus provides native support and specialized enhancement for LRM and LMM training workloads. Experimental results demonstrate significant performance gains in real-world ML scenarios. Magnus has been deployed at ByteDance for over five years, enabling robust and efficient data infrastructure for large-scale ML workloads.

At ByteDance, where we execute over a million Spark jobs and handle 500PB of shuffled data daily, ensuring resource efficiency is paramount for cost savings. However, achieving optimization of resource efficiency in large-scale production environments poses significant challenges. Drawing from our practical experiences, we have identified three key issues critical to addressing resource efficiency in real-world production settings: 1 slow I/Os leading to excessive CPU and memory idleness, 2 coarse-grained resource control causing wastage, and 3 sub-optimal job configurations resulting in low utilization. To tackle these issues, we propose a resource efficiency governance framework for Spark workloads. Specifically, 1 we devise the multi-mechanism shuffle services, including Enhanced External Shuffle Service (ESS) and Cloud Shuffle Service (CSS), where CSS employs a push-based approach to enhance I/O efficiency through sequential reading. 2 We modify the Spark configuration parameter protocol, allowing for fine-grained resource control by introducing several new parameters such as milliCores and memoryBurst, as well as supporting operators with additional spill modes. 3 We design a two-stage configuration autotuning method, comprising rule-based and algorithm-based tuning, providing more reliable Spark configuration optimizations. By deploying these techniques on millions of Spark jobs in production over the last two years, we have achieved over 22% CPU utilization increase, 5% memory utilization increase, and 10% shuffle block time ratio decrease, effectively saving millions of CPU cores and petabytes of memory daily.

In this paper, History-based Motion Vector Prediction (HMVP) is presented for video coding. With the proposed method, a table of HMVP candidates is maintained and updated on-the-fly. After decoding one inter-coded block, the table is updated by appending the associated motion information to the table as a new HMVP candidate. A First-In-First-Out (FIFO) rule is applied to manage the table. The HMVP candidates could be added to the Advanced Motion Vector Prediction (AMVP) candidate list as additional motion vector predictors. And they could also be added to the merge candidate list as additional merge candidates. With the proposed method, the motion information of previously coded blocks even not adjacent to the current block can be utilized for more efficient motion vector prediction. Simulation results have validated the efficiency of HMVP, wherein up to 4% BD rate saving could be achieved. The proposed method has been adopted by the next generation video coding standard, named Versatile Video Coding (VVC) developed by Joint Video Exploration Team (JVET).