More Like this

1. FlexPushdownDB: Hybrid Pushdown and Caching in a Cloud DBMS

   https://doi.org/10.14778/3476249.3476265  

   Yang, Yifei; Youill, Matt; Woicik, Matthew; Liu, Yizhou; Yu, Xiangyao; Serafini, Marco; Aboulnaga, Ashraf; Stonebraker, Michael (July 2021, Proceedings of the VLDB Endowment)

   null (Ed.)

   Modern cloud databases adopt a storage-disaggregation architecture that separates the management of computation and storage. A major bottleneck in such an architecture is the network connecting the computation and storage layers. Two solutions have been explored to mitigate the bottleneck: caching and computation pushdown. While both techniques can significantly reduce network traffic, existing DBMSs consider them as orthogonal techniques and support only one or the other, leaving potential performance benefits unexploited. In this paper we present FlexPushdownDB (FPDB), an OLAP cloud DBMS prototype that supports fine-grained hybrid query execution to combine the benefits of caching and computation pushdown in a storage-disaggregation architecture. We build a hybrid query executor based on a new concept called separable operators to combine the data from the cache and results from the pushdown processing. We also propose a novel Weighted-LFU cache replacement policy that takes into account the cost of pushdown computation. Our experimental evaluation on the Star Schema Benchmark shows that the hybrid execution outperforms both the conventional caching- only architecture and pushdown-only architecture by 2.2×. In the hybrid architecture, our experiments show that Weighted-LFU can outperform the baseline LFU by 37%. 

   more » « less
2. FlexpushdownDB: rethinking computation pushdown for cloud OLAP DBMSs

   https://doi.org/10.1007/s00778-024-00867-8  

   Yang, Yifei; Yu, Xiangyao; Serafini, Marco; Aboulnaga, Ashraf; Stonebraker, Michael (September 2024, The VLDB Journal)

   Modern cloud-native OLAP databases adopt a storage-disaggregation architecture that separates the management of compu- tation and storage. A major bottleneck in such an architecture is the network connecting the computation and storage layers. Computation pushdown is a promising solution to tackle this issue, which offloads some computation tasks to the storage layer to reduce network traffic. This paper presents FlexPushdownDB (FPDB), where we revisit the design of computation pushdown in a storage-disaggregation architecture, and then introduce several optimizations to further accelerate query pro- cessing. First, FPDB supports hybrid query execution, which combines local computation on cached data and computation pushdown to cloud storage at a fine granularity. Within the cache, FPDB uses a novel Weighted-LFU cache replacement policy that takes into account the cost of pushdown computation. Second, we design adaptive pushdown as a new mecha- nism to avoid throttling the storage-layer computation during pushdown, which pushes the request back to the computation layer at runtime if the storage-layer computational resource is insufficient. Finally, we derive a general principle to identify pushdown-amenable computational tasks, by summarizing common patterns of pushdown capabilities in existing systems, and further propose two new pushdown operators, namely, selection bitmap and distributed data shuffle. Evaluation on SSB and TPC-H shows each optimization can improve the performance by 2.2×, 1.9×, and 3× respectively. 

   more » « less
3. Distributed VLMs: Efficient vision-language processing through cloud-edge collaboration

   Li, Y; Gumaste, D; Turkcan, M; Ghaderi, LJ; Zussman, G; Kostic, Z (February 2025, in Proc. 4th IEEE Workshop on Pervasive and Resource-constrained Artificial Intelligence (PeRConAI), 2025.)

   Vision Language models (VLMs) have transformed Generative AI by enabling systems to interpret and respond to multi-modal data in real-time. While advancements in edge computing have made it possible to deploy smaller Large Language Models (LLMs) on smartphones and laptops, deploying competent VLMs on edge devices remains challenging due to their high computational demands. Furthermore, cloud-only deployments fail to utilize the evolving processing capabilities at the edge and limit responsiveness. This paper introduces a distributed architecture for VLMs that addresses these limitations by partitioning model components between edge devices and central servers. In this setup, vision components run on edge devices for immediate processing, while language generation of the VLM is handled by a centralized server, resulting in up to 33% improvement in throughput over traditional cloud-only solutions. Moreover, our approach enhances the computational efficiency of off-the-shelf VLM models without the need for model compression techniques. This work demonstrates the scalability and efficiency of a hybrid architecture for VLM deployment and contributes to the discussion on how distributed approaches can improve VLM performance. Index Terms—vision-language models (VLMs), edge computing, distributed computing, inference optimization, edge-cloud collaboration. 

   more » « less
4. A 99.7% Efficient 300 W Hard Disk Drive Storage Server with Multiport Ac-Coupled Differential Power Processing (MAC-DPP) Architecture

   https://doi.org/10.1109/ECCE.2019.8911853  

   Wang, Ping; Chen, Yenan; Kushima, Parker; Elasser, Youssef; Liu, Ming; Chen, Minjie (September 2019, 2019 IEEE Energy Conversion Congress and Exposition (ECCE))

   High performance computing needs high performance power electronics. This paper presents the design of an ultra-efficient series-stacked hard-disk-drive (HDD) data storage server with a multiport ac-coupled differential power processing (MAC-DPP) architecture. A large number of HDDs are connected in series and ac-coupled through a multi-winding transformer with a single flux linkage. The MAC-DPP architecture offers very low power conversion stress, can achieve extremely high efficiency, and can reduce the magnetic size and the component count. A hybrid time-sharing and distributed phase-shift control strategy is developed to modulate the ac-coupled multi-input-multi-output (MIMO) power flow. A 10-port MAC-DPP prototype was designed to support a 300 W data storage system with 10 series-stacked voltage domains. The MAC-DPP converter was tested with a 50-HDD 12TB testbench, which can maintain normal operation of the server against the worst hot-swapping scenario. The 300 W MAC-DPP prototype can achieve 99.7% peak system efficiency and over 100 W/in 3 power density. 

   more » « less
5. Domain Isolation in FPGA-Accelerated Cloud and Data Center Applications

   https://doi.org/10.1145/3453688.3461527  

   Mandebi Mbongue, Joel; Saha, Sujan Kumar; Bobda, Christophe (June 2021, Proceedings of the 2021 on Great Lakes Symposium on VLSI. 2021)

   null (Ed.)

   Cloud and data center applications increasingly leverage FPGAs because of their performance/watt benefits and flexibility advantages over traditional processing cores such as CPUs and GPUs. As the rising demand for hardware acceleration gradually leads to FPGA multi-tenancy in the cloud, there are rising concerns about the security challenges posed by FPGA virtualization. Exposing space-shared FPGAs to multiple cloud tenants may compromise the confidentiality, integrity, and availability of FPGA-accelerated applications. In this work, we present a hardware/software architecture for domain isolation in FPGA-accelerated clouds and data centers with a focus on software-based attacks aiming at unauthorized access and information leakage. Our proposed architecture implements Mandatory Access Control security policies from software down to the hardware accelerators on FPGA. Our experiments demonstrate that the proposed architecture protects against such attacks with minimal area and communication overhead. 

   more » « less