TASM: A Tile-Based Storage Manager for Video Analytics

Daum, Maureen; Haynes, Brandon; He, Duanwei; Mazumdar, Amrita; Balazinska, Magdalena

doi:10.48550/arxiv.2006.02958

Cited by 4 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TASM [8] is a storage manager for video data that improves video query performance. TASM speeds up queries that retrieve objects in a video with low storage overhead and good video quality by splitting the video frames into independent tiles and optimizes the video file layout based on its content and the query workload.…”

Section: Tasmmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems

Ibrahim,

Maurya,

Jafari

et al. 2021

Preprint

View full text Add to dashboard Cite

Video analytics systems perform automatic events, movements, and actions recognition in a video and make it possible to execute queries on the video. As a result of a large number of video data that need to be processed, optimizing the performance of video analytics systems has become an important research topic. Neural networks are the state-of-the-art for performing video analytics tasks such as video annotation and object detection. Prior survey papers consider application-specific video analytics techniques that improve accuracy of the results; however, in this survey paper, we provide a review of the techniques that focus on optimizing the performance of Neural Network-Based Video Analytics Systems.

show abstract

Section: Tasmmentioning

confidence: 99%

“…Applications use video analytics to take immediate actions based on their decision without any human interactions. Besides, several database management systems that work with data and query processing on videos have more advanced features [8] to support video analytics.…”

Section: Introductionmentioning

confidence: 99%

A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems

Ibrahim,

Maurya,

Jafari

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…To do so, the application uses a pipeline to detect frames with people and/or cars, and match them to specific individual's faces and car descriptions, respectively. Another example is an ornithologist (a scientist who studies birds), who is looking to study a specific bird species, will deploy multiple cameras where the birds may reside or eat [29,45]. To analyze the feed, the ornithologist can build a pipeline to detect the presence of birds, classify detected birds into specific bird species, and enhance frames corresponding to specific species for further study.…”

Section: Introductionmentioning

confidence: 99%

“…There have also been several frameworks enabling users to run general analytics dataflow and query execution [32,40,54,55,73], as well as proposed optimizations and techniques for executing these framework's workflows on a broad range of execution engines [37] and hardware platforms [58]. These frameworks are not well-suited for the needs of dominating and emerging video analytics operations such as encoding and decoding videos [35,57], trading off cost for both latency and quality [42,74], and querying large video datasets [29,39]. They also rely on users to properly allocate the right amount of resources and configure operation knobs to meet diverse application targets.…”

Section: Introductionmentioning

confidence: 99%

Llama: A Heterogeneous & Serverless Framework for Auto-Tuning Video Analytics Pipelines

Romero,

Zhao,

Yadwadkar

et al. 2021

Preprint

View full text Add to dashboard Cite

The proliferation of camera-enabled devices and large video repositories has given rise to a diverse set of video analytics applications. The video pipelines for these applications are DAGs of operations that transform videos, process extracted metadata, and answer questions such as, "Is this intersection congested?" The latency and resource efficiency of pipelines can be optimized using configurable knobs for each operation such as the sampling rate, batch size, or type of hardware used. However, determining efficient configurations is challenging because (a) the configuration search space is exponentially large, and (b) the optimal configuration depends on the desired latency target and the input video contents that may exercise different paths in the DAG and produce different volumes of intermediate results. Existing video analytics and processing systems leave it to the users to manually configure operations and select hardware resources. Hence, we observe that they often execute inefficiently and fail to meet latency and cost targets.We present Llama: a heterogeneous and serverless framework for auto-tuning video pipelines. Llama optimizes the overall video pipeline latency by (a) dynamically calculating latency targets peroperation invocation, and (b) dynamically running a cost-based optimizer to determine efficient configurations that meet the target latency for each invocation. This makes the problem of auto-tuning large video pipelines tractable and allows us to handle input dependent behavior, conditional branches in the DAG, and execution variability. It also allows us to dynamically target heterogenous hardware and minimize execution cost. We describe the algorithms in Llama and evaluate it on a cloud platform using serverless CPU and GPU resources. We show that compared to state-of-the-art cluster and serverless video analytics and processing systems, Llama achieves 7.9× lower latency and 17.2× cost reduction on average.

show abstract

“…[18,42]). In practice, of course, videos are stored on disk, and the cost of reading and decompressing is high relative to subsequent processing [11,18], e.g., constituting more than 50% of total runtime [28]. The result is a performance plateau limited by Amdahl's law, where an emphasis on post-decompression performance might yield impressive results in isolation, but ignores the diminishing returns when performance is evaluated end-to-end.…”

Section: Introductionmentioning

confidence: 99%

VSS: A Storage System for Video Analytics [Technical Report]

Haynes,

Daum,

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We present a new video storage system (VSS) designed to decouple high-level video operations from the low-level details required to store and efficiently retrieve video data. VSS is designed to be the storage subsystem of a video data management system (VDBMS) and is responsible for:(1) transparently and automatically arranging the data on disk in an efficient, granular format; (2) caching frequentlyretrieved regions in the most useful formats; and (3) eliminating redundancies found in videos captured from multiple cameras with overlapping fields of view. Our results suggest that VSS can improve VDBMS read performance by up to 54%, reduce storage costs by up to 45%, and enable developers to focus on application logic rather than video storage and retrieval.

show abstract

TASM: A Tile-Based Storage Manager for Video Analytics

Cited by 4 publications

References 0 publications

A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems

A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems

Llama: A Heterogeneous & Serverless Framework for Auto-Tuning Video Analytics Pipelines

VSS: A Storage System for Video Analytics [Technical Report]

Contact Info

Product

Resources

About