Pub Date : 2023-08-01DOI: 10.14778/3611540.3611591
Zhuo Ma, Yilong Yang, Bin Xiao, Yang Liu, Xinjing Liu, Zhuoran Ma, Tong Yang
Recent works explore several attacks against Machine-Learning-as-a-Service (MLaaS) platforms (e.g., the model stealing attack), allegedly posing potential real-world threats beyond viability in laboratories. However, hampered by model-type-sensitive , most of the attacks can hardly break mainstream real-world MLaaS platforms. That is, many MLaaS attacks are designed against only one certain type of model, such as tree models or neural networks. As the black-box MLaaS interface hides model type info, the attacker cannot choose a proper attack method with confidence, limiting the attack performance. In this paper, we demonstrate a system, named Sniffer, that is capable of making model-type-sensitive attacks "great again" in real-world applications. Specifically, Sniffer consists of four components: Generator, Querier, Probe, and Arsenal. The first two components work for preparing attack samples. Probe, as the most characteristic component in Sniffer, implements a series of self-designed algorithms to determine the type of models hidden behind the black-box MLaaS interfaces. With model type info unraveled, an optimum method can be selected from Arsenal (containing multiple attack methods) to accomplish its attack. Our demonstration shows how the audience can interact with Sniffer in a web-based interface against five mainstream MLaaS platforms.
{"title":"Sniffer: A Novel Model Type Detection System against Machine-Learning-as-a-Service Platforms","authors":"Zhuo Ma, Yilong Yang, Bin Xiao, Yang Liu, Xinjing Liu, Zhuoran Ma, Tong Yang","doi":"10.14778/3611540.3611591","DOIUrl":"https://doi.org/10.14778/3611540.3611591","url":null,"abstract":"Recent works explore several attacks against Machine-Learning-as-a-Service (MLaaS) platforms (e.g., the model stealing attack), allegedly posing potential real-world threats beyond viability in laboratories. However, hampered by model-type-sensitive , most of the attacks can hardly break mainstream real-world MLaaS platforms. That is, many MLaaS attacks are designed against only one certain type of model, such as tree models or neural networks. As the black-box MLaaS interface hides model type info, the attacker cannot choose a proper attack method with confidence, limiting the attack performance. In this paper, we demonstrate a system, named Sniffer, that is capable of making model-type-sensitive attacks \"great again\" in real-world applications. Specifically, Sniffer consists of four components: Generator, Querier, Probe, and Arsenal. The first two components work for preparing attack samples. Probe, as the most characteristic component in Sniffer, implements a series of self-designed algorithms to determine the type of models hidden behind the black-box MLaaS interfaces. With model type info unraveled, an optimum method can be selected from Arsenal (containing multiple attack methods) to accomplish its attack. Our demonstration shows how the audience can interact with Sniffer in a web-based interface against five mainstream MLaaS platforms.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134998300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611553
Liang Lin, Yuhan Li, Bin Wu, Huijun Mai, Renjie Lou, Jian Tan, Feifei Li
The surge in data analytics has fostered burgeoning demand for AnalyticDB on Alibaba Cloud, which has well served thousands of customers from various business sectors. The most notable feature is the diversity of the workloads it handles, including batch processing, real-time data analytics, and unstructured data analytics. To improve the overall performance for such diverse workloads, one of the major challenges is to optimize long-running complex queries without sacrificing the processing efficiency of short-running interactive queries. While existing methods attempt to utilize runtime dynamic statistics for adaptive query processing, they often focus on specific scenarios instead of providing a holistic solution. To address this challenge, we propose a new framework called Anser , which enhances the design of traditional distributed data warehouses by embedding a new information sharing mechanism. This allows for the efficient management of the production and consumption of various dynamic information across the system. Building on top of Anser , we introduce a novel scheduling policy that optimizes both data and information exchanges within the physical plan, enabling the acceleration of complex analytical queries without sacrificing the performance of short-running interactive queries. We conduct comprehensive experiments over public and in-house workloads to demonstrate the effectiveness and efficiency of our proposed information sharing framework.
{"title":"Anser: Adaptive Information Sharing Framework of AnalyticDB","authors":"Liang Lin, Yuhan Li, Bin Wu, Huijun Mai, Renjie Lou, Jian Tan, Feifei Li","doi":"10.14778/3611540.3611553","DOIUrl":"https://doi.org/10.14778/3611540.3611553","url":null,"abstract":"The surge in data analytics has fostered burgeoning demand for AnalyticDB on Alibaba Cloud, which has well served thousands of customers from various business sectors. The most notable feature is the diversity of the workloads it handles, including batch processing, real-time data analytics, and unstructured data analytics. To improve the overall performance for such diverse workloads, one of the major challenges is to optimize long-running complex queries without sacrificing the processing efficiency of short-running interactive queries. While existing methods attempt to utilize runtime dynamic statistics for adaptive query processing, they often focus on specific scenarios instead of providing a holistic solution. To address this challenge, we propose a new framework called Anser , which enhances the design of traditional distributed data warehouses by embedding a new information sharing mechanism. This allows for the efficient management of the production and consumption of various dynamic information across the system. Building on top of Anser , we introduce a novel scheduling policy that optimizes both data and information exchanges within the physical plan, enabling the acceleration of complex analytical queries without sacrificing the performance of short-running interactive queries. We conduct comprehensive experiments over public and in-house workloads to demonstrate the effectiveness and efficiency of our proposed information sharing framework.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135003293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611607
Yuchen Peng, Ke Chen, Lidan Shou, Dawei Jiang, Gang Chen
Data analysts nowadays are keen to have analytical capabilities involving deep learning (DL). Collaborative queries, which employ relational operations to process structured data and DL models to process unstructured data, provide a powerful facility for DL-based in-database analysis. The classical approach to support collaborative queries in relational databases is to integrate DL models with user-defined functions (UDFs) in a general-purpose language (e.g., C++) to process unstructured data. This approach suffers from suboptimal performance as the opaque UDFs preclude the generation of an optimal query plan. A recent work, DL2SQL, addresses the problem of collaborative query optimization by first converting DL computations into SQL subqueries and then using a classical relational query optimizer to optimize the entire collaborative query. However, the DL2SQL approach compromises usability by requiring data analysts to manually manage DL-related data and tune query performance. To this end, this paper introduces AQUA, an analytical database designed for efficient collaborative query processing. Built on DL2SQL, AQUA automates translations from collaborative queries into SQL queries. To enhance usability, AQUA introduces two techniques: 1) a declarative scheme for DL-related data management, and 2) DL-specific optimizations for collaborative query processing, eliminating the burden of manual data management and performance tuning from the data analysts. We demonstrate the key contributions of AQUA via a web APP that allows the audience to perform collaborative queries on the CIFAR-10 dataset.
{"title":"AQUA: Automatic Collaborative Query Processing in Analytical Database","authors":"Yuchen Peng, Ke Chen, Lidan Shou, Dawei Jiang, Gang Chen","doi":"10.14778/3611540.3611607","DOIUrl":"https://doi.org/10.14778/3611540.3611607","url":null,"abstract":"Data analysts nowadays are keen to have analytical capabilities involving deep learning (DL). Collaborative queries, which employ relational operations to process structured data and DL models to process unstructured data, provide a powerful facility for DL-based in-database analysis. The classical approach to support collaborative queries in relational databases is to integrate DL models with user-defined functions (UDFs) in a general-purpose language (e.g., C++) to process unstructured data. This approach suffers from suboptimal performance as the opaque UDFs preclude the generation of an optimal query plan. A recent work, DL2SQL, addresses the problem of collaborative query optimization by first converting DL computations into SQL subqueries and then using a classical relational query optimizer to optimize the entire collaborative query. However, the DL2SQL approach compromises usability by requiring data analysts to manually manage DL-related data and tune query performance. To this end, this paper introduces AQUA, an analytical database designed for efficient collaborative query processing. Built on DL2SQL, AQUA automates translations from collaborative queries into SQL queries. To enhance usability, AQUA introduces two techniques: 1) a declarative scheme for DL-related data management, and 2) DL-specific optimizations for collaborative query processing, eliminating the burden of manual data management and performance tuning from the data analysts. We demonstrate the key contributions of AQUA via a web APP that allows the audience to perform collaborative queries on the CIFAR-10 dataset.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135003650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611542
Alex Depoutovitch, Chong Chen, Per-Ake Larson, Jack Ng, Shu Lin, Guanzhu Xiong, Paul Lee, Emad Boctor, Samiao Ren, Lengdong Wu, Yuchen Zhang, Calvin Sun
A single-master database has limited update capacity because a single node handles all updates. A multi-master database potentially has higher update capacity because the load is spread across multiple nodes. However, the need to coordinate updates and ensure durability can generate high network traffic. Reducing network load is particularly important in a cloud environment where the network infrastructure is shared among thousands of tenants. In this paper, we present Taurus MM, a shared-storage multi-master database optimized for cloud environments. It implements two novel algorithms aimed at reducing network traffic plus a number of additional optimizations. The first algorithm is a new type of distributed clock that combines the small size of Lamport clocks with the effective support of distributed snapshots of vector clocks. The second algorithm is a new hybrid page and row locking protocol that significantly reduces the number of lock requests sent over the network. Experimental results on a cluster with up to eight masters demonstrate superior performance compared to Aurora multi-master and CockroachDB.
{"title":"Taurus MM: Bringing Multi-Master to the Cloud","authors":"Alex Depoutovitch, Chong Chen, Per-Ake Larson, Jack Ng, Shu Lin, Guanzhu Xiong, Paul Lee, Emad Boctor, Samiao Ren, Lengdong Wu, Yuchen Zhang, Calvin Sun","doi":"10.14778/3611540.3611542","DOIUrl":"https://doi.org/10.14778/3611540.3611542","url":null,"abstract":"A single-master database has limited update capacity because a single node handles all updates. A multi-master database potentially has higher update capacity because the load is spread across multiple nodes. However, the need to coordinate updates and ensure durability can generate high network traffic. Reducing network load is particularly important in a cloud environment where the network infrastructure is shared among thousands of tenants. In this paper, we present Taurus MM, a shared-storage multi-master database optimized for cloud environments. It implements two novel algorithms aimed at reducing network traffic plus a number of additional optimizations. The first algorithm is a new type of distributed clock that combines the small size of Lamport clocks with the effective support of distributed snapshots of vector clocks. The second algorithm is a new hybrid page and row locking protocol that significantly reduces the number of lock requests sent over the network. Experimental results on a cluster with up to eight masters demonstrate superior performance compared to Aurora multi-master and CockroachDB.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135003928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611605
Chaozu Zhang, Qiaomu Shen, Bo Tang
Nowadays, Apache Hive has been widely used for large-scale data analysis applications in many organizations. Various visual analytical tools are developed to help Hive users quickly analyze the query execution process and identify the performance bottleneck of executed queries. However, existing tools mostly focus on showing the time usage of query sub-components (jobs and operators) but fail to provide enough evidence to analyze the root reasons for the slow execution progress. To tackle this problem, we develop a visual analytical system DHive to visualize and analyze the query execution progress via dataflow analysis. DHive shows the dataflow during query execution at multiple levels: query level, job level and task level, which enable users to identify the key jobs/tasks and explain their time usage by linking them to the auxiliary information such as the system configuration and hardware status. We demonstrate the effectiveness of DHive by two cases in a production cluster. DHive is open-source at https://github.com/DBGroup-SUSTech/DHive.git.
{"title":"DHive: Query Execution Performance Analysis via Dataflow in Apache Hive","authors":"Chaozu Zhang, Qiaomu Shen, Bo Tang","doi":"10.14778/3611540.3611605","DOIUrl":"https://doi.org/10.14778/3611540.3611605","url":null,"abstract":"Nowadays, Apache Hive has been widely used for large-scale data analysis applications in many organizations. Various visual analytical tools are developed to help Hive users quickly analyze the query execution process and identify the performance bottleneck of executed queries. However, existing tools mostly focus on showing the time usage of query sub-components (jobs and operators) but fail to provide enough evidence to analyze the root reasons for the slow execution progress. To tackle this problem, we develop a visual analytical system DHive to visualize and analyze the query execution progress via dataflow analysis. DHive shows the dataflow during query execution at multiple levels: query level, job level and task level, which enable users to identify the key jobs/tasks and explain their time usage by linking them to the auxiliary information such as the system configuration and hardware status. We demonstrate the effectiveness of DHive by two cases in a production cluster. DHive is open-source at https://github.com/DBGroup-SUSTech/DHive.git.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"222 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134998307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611634
Alon Halevy, Yejin Choi, Avrilia Floratou, Michael J. Franklin, Natasha Noy, Haixun Wang
Large language models (LLMs) have recently taken the world by storm, promising potentially game changing opportunities in multiple fields. Naturally, there is significant promise in applying LLMs to the management of structured data, or more generally, to the processes involved in data science. At the very least, LLMs have the potential to provide substantial advancements in long-standing challenges that our community has been tackling for decades. On the other hand, they may introduce completely new capabilities that we have only dreamed of thus far. This panel will bring together a few leading experts who have been thinking about these opportunities from various perspectives and fielding them in research prototypes and even in commercial applications.
{"title":"Will LLMs Reshape, Supercharge, or Kill Data Science? (VLDB 2023 Panel)","authors":"Alon Halevy, Yejin Choi, Avrilia Floratou, Michael J. Franklin, Natasha Noy, Haixun Wang","doi":"10.14778/3611540.3611634","DOIUrl":"https://doi.org/10.14778/3611540.3611634","url":null,"abstract":"Large language models (LLMs) have recently taken the world by storm, promising potentially game changing opportunities in multiple fields. Naturally, there is significant promise in applying LLMs to the management of structured data, or more generally, to the processes involved in data science. At the very least, LLMs have the potential to provide substantial advancements in long-standing challenges that our community has been tackling for decades. On the other hand, they may introduce completely new capabilities that we have only dreamed of thus far. This panel will bring together a few leading experts who have been thinking about these opportunities from various perspectives and fielding them in research prototypes and even in commercial applications.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134998128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611610
Nico Schäfer, Damjan Gjurovski, Angjela Davitkova, Sebastian Michel
While existing data management solutions try to keep up with novel data formats and features, a myriad of valuable functionality is often only accessible via programming language libraries. Particularly for machine learning tasks, there is a wealth of pre-trained models and easy-to-use libraries that allow a wide audience to harness state-of-the-art machine learning. We propose the demonstration of a highly modularized data processor for semi-structured data that can be extended by means of plain Python scripts. Next to commonly supported user-defined functions, the deep decomposition allows augmenting the core engine with additional index structures, customized import and export routines, and custom aggregation functions. For several use cases, we detail how user-defined modules can be quickly realized and invite the audience to write and apply custom code, to tailor provided code snippets that we bring along to own preferences to solve data analytics tasks involving sentiment analysis of Twitter tweets.
{"title":"To UDFs and Beyond: Demonstration of a Fully Decomposed Data Processor for General Data Wrangling Tasks","authors":"Nico Schäfer, Damjan Gjurovski, Angjela Davitkova, Sebastian Michel","doi":"10.14778/3611540.3611610","DOIUrl":"https://doi.org/10.14778/3611540.3611610","url":null,"abstract":"While existing data management solutions try to keep up with novel data formats and features, a myriad of valuable functionality is often only accessible via programming language libraries. Particularly for machine learning tasks, there is a wealth of pre-trained models and easy-to-use libraries that allow a wide audience to harness state-of-the-art machine learning. We propose the demonstration of a highly modularized data processor for semi-structured data that can be extended by means of plain Python scripts. Next to commonly supported user-defined functions, the deep decomposition allows augmenting the core engine with additional index structures, customized import and export routines, and custom aggregation functions. For several use cases, we detail how user-defined modules can be quickly realized and invite the audience to write and apply custom code, to tailor provided code snippets that we bring along to own preferences to solve data analytics tasks involving sentiment analysis of Twitter tweets.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134996890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611621
Chen Qian, Shiwei Liang, Zhaoyang Wang, Yin Lou
In many applications the training data do not always contain sufficient information to produce high-quality decision rules for standard (end-to-end) rule mining algorithms, and human experts have to incorporate domain knowledge during rule induction in order to get meaningful results. In this work we present Fanglue, a home-grown system inside Alipay, for interactive decision rule crafting. Fanglue is a distributed in-memory system and is highly responsive when processing large-scale datasets. In addition, Fanglue extends the standard representation of a decision rule by introducing disjunctive clauses. Having disjunctive clauses can improve the coverage and robustness of a decision rule, especially for fraud prevention in Fintech applications.
{"title":"Fanglue: An Interactive System for Decision Rule Crafting","authors":"Chen Qian, Shiwei Liang, Zhaoyang Wang, Yin Lou","doi":"10.14778/3611540.3611621","DOIUrl":"https://doi.org/10.14778/3611540.3611621","url":null,"abstract":"In many applications the training data do not always contain sufficient information to produce high-quality decision rules for standard (end-to-end) rule mining algorithms, and human experts have to incorporate domain knowledge during rule induction in order to get meaningful results. In this work we present Fanglue, a home-grown system inside Alipay, for interactive decision rule crafting. Fanglue is a distributed in-memory system and is highly responsive when processing large-scale datasets. In addition, Fanglue extends the standard representation of a decision rule by introducing disjunctive clauses. Having disjunctive clauses can improve the coverage and robustness of a decision rule, especially for fraud prevention in Fintech applications.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134997928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611566
Zhicheng Pan, Yihang Wang, Yingying Zhang, Sean Bin Yang, Yunyao Cheng, Peng Chen, Chenjuan Guo, Qingsong Wen, Xiduo Tian, Yunliang Dou, Zhiqiang Zhou, Chengcheng Yang, Aoying Zhou, Bin Yang
Predictive autoscaling is a key enabler for optimizing cloud resource allocation in Alibaba Cloud's computing platforms, which dynamically adjust the Elastic Compute Service (ECS) instances based on predicted user demands to ensure Quality of Service (QoS). However, user demands in the cloud are often highly complex, with high uncertainty and scale-sensitive temporal dependencies, thus posing great challenges for accurate prediction of future demands. These in turn make autoscaling challenging---autoscaling needs to properly account for demand uncertainty while maintaining a reasonable trade-off between two contradictory factors, i.e., low instance running costs vs. low QoS violation risks. To address the above challenges, we propose a novel predictive autoscaling framework MagicScaler , consisting of a Multi-scale attentive Gaussian process based predictor and an uncertainty-aware scaler. First, the predictor carefully bridges the best of two successful prediction methodologies---multi-scale attention mechanisms, which are good at capturing complex, multi-scale features, and stochastic process regression, which can quantify prediction uncertainty, thus achieving accurate demand prediction with quantified uncertainty. Second, the scaler takes the quantified future demand uncertainty into a judiciously designed loss function with stochastic constraints, enabling flexible trade-off between running costs and QoS violation risks. Extensive experiments on three clusters of Alibaba Cloud in different Chinese cities demonstrate the effectiveness and efficiency of MagicScaler , which outperforms other commonly adopted scalers, thus justifying our design choices.
{"title":"MagicScaler: Uncertainty-Aware, Predictive Autoscaling","authors":"Zhicheng Pan, Yihang Wang, Yingying Zhang, Sean Bin Yang, Yunyao Cheng, Peng Chen, Chenjuan Guo, Qingsong Wen, Xiduo Tian, Yunliang Dou, Zhiqiang Zhou, Chengcheng Yang, Aoying Zhou, Bin Yang","doi":"10.14778/3611540.3611566","DOIUrl":"https://doi.org/10.14778/3611540.3611566","url":null,"abstract":"Predictive autoscaling is a key enabler for optimizing cloud resource allocation in Alibaba Cloud's computing platforms, which dynamically adjust the Elastic Compute Service (ECS) instances based on predicted user demands to ensure Quality of Service (QoS). However, user demands in the cloud are often highly complex, with high uncertainty and scale-sensitive temporal dependencies, thus posing great challenges for accurate prediction of future demands. These in turn make autoscaling challenging---autoscaling needs to properly account for demand uncertainty while maintaining a reasonable trade-off between two contradictory factors, i.e., low instance running costs vs. low QoS violation risks. To address the above challenges, we propose a novel predictive autoscaling framework MagicScaler , consisting of a Multi-scale attentive Gaussian process based predictor and an uncertainty-aware scaler. First, the predictor carefully bridges the best of two successful prediction methodologies---multi-scale attention mechanisms, which are good at capturing complex, multi-scale features, and stochastic process regression, which can quantify prediction uncertainty, thus achieving accurate demand prediction with quantified uncertainty. Second, the scaler takes the quantified future demand uncertainty into a judiciously designed loss function with stochastic constraints, enabling flexible trade-off between running costs and QoS violation risks. Extensive experiments on three clusters of Alibaba Cloud in different Chinese cities demonstrate the effectiveness and efficiency of MagicScaler , which outperforms other commonly adopted scalers, thus justifying our design choices.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134998134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.14778/3611540.3611638
Yejin Choi
Scale appears to be the winning recipe in today's leaderboards. And yet, extreme-scale neural models are (un)surprisingly brittle and make errors that are often nonsensical and even counterintuitive. In this talk, I will argue for the importance of knowledge, especially commonsense knowledge, as well as inference-time reasoning algorithms, and demonstrate how smaller models developed in academia can still have an edge over larger industry-scale models, if powered with knowledge and/or reasoning algorithms.
{"title":"Common Sense: The Dark Matter of Language and Intelligence (VLDB 2023 Keynote)","authors":"Yejin Choi","doi":"10.14778/3611540.3611638","DOIUrl":"https://doi.org/10.14778/3611540.3611638","url":null,"abstract":"Scale appears to be the winning recipe in today's leaderboards. And yet, extreme-scale neural models are (un)surprisingly brittle and make errors that are often nonsensical and even counterintuitive. In this talk, I will argue for the importance of knowledge, especially commonsense knowledge, as well as inference-time reasoning algorithms, and demonstrate how smaller models developed in academia can still have an edge over larger industry-scale models, if powered with knowledge and/or reasoning algorithms.","PeriodicalId":54220,"journal":{"name":"Proceedings of the Vldb Endowment","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135002982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：