Jian Tan
Jian Tan has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Bin Wu
Alibaba Cloud Computing, Hangzhou, China
3 shared publications - Ye Li
Alibaba Cloud Computing, Hangzhou, China
3 shared publications - Feifei Li
Alibaba Cloud, Hangzhou, China
2 shared publications - Ju Fan
Renmin University of China
2 shared publications - Nan Tang
The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China
2 shared publications - Pengyi Wang
Renmin University of China, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 243-246
A ndromeda : Debugging Database Performance Issues with Retrieval-Augmented Large Language Models
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 243-246
Open AccessA ndromeda : Debugging Database Performance Issues with Retrieval-Augmented Large Language Models
- Open Access
- 10 February 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (1) , 1-27
Automatic Database Configuration Debugging using Retrieval-Augmented Language Models
Database management system (DBMS) configuration debugging, e.g., diagnosing poorly configured DBMS knobs and generating troubleshooting recommendations, is crucial in optimizing DBMS performance. However, the configuration debugging process is tedious and, sometimes challenging, even for seasoned database administrators (DBAs) with sufficient experience in DBMS configurations and good understandings of the DBMS internals (e.g., MySQL or Oracle). To address this difficulty, we propose Andromeda, a framework that utilizes large language models (LLMs) to enable automatic DBMS configuration debugging. Andromeda serves as a natural surrogate of DBAs to answer a wide range of natural language (NL) questions on DBMS configuration issues, and to generate diagnostic suggestions to fix these issues. Nevertheless, directly prompting LLMs with these professional questions may result in overly generic and often unsatisfying answers. To this end, we propose a retrieval-augmented generation (RAG) strategy that effectively provides matched domain-specific contexts for the question from multiple sources. They come from related historical questions, troubleshooting manuals and DBMS telemetries, which significantly improve the performance of configuration debugging. To support the RAG strategy, we develop a document retrieval mechanism addressing heterogeneous documents and design an effective method for telemetry analysis. Extensive experiments on real-world DBMS configuration debugging datasets show that Andromeda significantly outperforms existing solutions.
- 10 February 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (1) , 1-27
Open AccessAutomatic Database Configuration Debugging using Retrieval-Augmented Language Models
Database management system (DBMS) configuration debugging, e.g., diagnosing poorly configured DBMS knobs and generating troubleshooting recommendations, is crucial in optimizing DBMS performance. However, the configuration debugging process is tedious and, sometimes challenging, even for seasoned database administrators (DBAs) with sufficient experience in DBMS configurations and good understandings of the DBMS internals (e.g., MySQL or Oracle). To address this difficulty, we propose Andromeda, a framework that utilizes large language models (LLMs) to enable automatic DBMS configuration debugging. Andromeda serves as a natural surrogate of DBAs to answer a wide range of natural language (NL) questions on DBMS configuration issues, and to generate diagnostic suggestions to fix these issues. Nevertheless, directly prompting LLMs with these professional questions may result in overly generic and often unsatisfying answers. To this end, we propose a retrieval-augmented generation (RAG) strategy that effectively provides matched domain-specific contexts for the question from multiple sources. They come from related historical questions, troubleshooting manuals and DBMS telemetries, which significantly improve the performance of configuration debugging. To support the RAG strategy, we develop a document retrieval mechanism addressing heterogeneous documents and design an effective method for telemetry analysis. Extensive experiments on real-world DBMS configuration debugging datasets show that Andromeda significantly outperforms existing solutions.
- Open Access
- 12 September 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Modeling and Performance Evaluation of Computing Systems
- Vol. 9 (4) , 1-24
Learning-Augmented Energy-Aware List Scheduling for Precedence-Constrained Tasks
We study the problem of scheduling precedence-constrained tasks to balance between performance and energy consumption. We consider a system with multiple servers capable of speed scaling and seek to schedule precedence-constrained tasks to minimize a linear combination of performance and energy consumption. Inspired by the single-server setting, we propose the concept of pseudo-size for individual tasks, which is a measure of the externalities of a task in the precedence graph and is learned from historical workload data. We then propose a two-stage scheduling framework that uses a learned pseudo-size approximation and achieves a provable approximation bound on the linear combination of performance and energy consumption for both makespan and total weighted completion time, where the quality of the bound depends on the approximation quality of pseudo-sizes. We show experimentally that learning-based approaches consistently perform near optimally.
- 12 September 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Modeling and Performance Evaluation of Computing Systems
- Vol. 9 (4) , 1-24
Open AccessLearning-Augmented Energy-Aware List Scheduling for Precedence-Constrained Tasks
We study the problem of scheduling precedence-constrained tasks to balance between performance and energy consumption. We consider a system with multiple servers capable of speed scaling and seek to schedule precedence-constrained tasks to minimize a linear combination of performance and energy consumption. Inspired by the single-server setting, we propose the concept of pseudo-size for individual tasks, which is a measure of the externalities of a task in the precedence graph and is learned from historical workload data. We then propose a two-stage scheduling framework that uses a learned pseudo-size approximation and achieves a provable approximation bound on the linear combination of performance and energy consumption for both makespan and total weighted completion time, where the quality of the bound depends on the approximation quality of pseudo-sizes. We show experimentally that learning-based approaches consistently perform near optimally.
- Open Access
- 13 June 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 1 (2) , 1-26
A Unified and Efficient Coordinating Framework for Autonomous DBMS Tuning
Recently using machine learning (ML) based techniques to optimize the performance of modern database management systems (DBMSs) has attracted intensive interest from both industry and academia. With an objective to tune a specific component of a DBMS (e.g., index selection, knobs tuning), the ML-based tuning agents have shown to be able to find better configurations than experienced database administrators (DBAs). However, one critical yet challenging question remains unexplored -- how to make those ML-based tuning agents work collaboratively. Existing methods do not consider the dependencies among the multiple agents, and the model used by each agent only studies the effect of changing the configurations in a single component. To tune different components for DBMS, a coordinating mechanism is needed to make the multiple agents be cognizant of each other. Also, we need to decide how to allocate the limited tuning budget (e.g., time and resources) among the agents to maximize the performance. Such a decision is difficult to make since the distribution of the reward (i.e., performance improvement) corresponding to each agent is unknown and non-stationary. In this paper, we study the above question and present a unified coordinating framework to efficiently utilize existing ML-based agents. First, we propose a message propagation protocol that specifies the collaboration behaviors for agents and encapsulates the global tuning messages in each agent's model. Second, we combine Thompson Sampling, a well-studied reinforcement learning algorithm with a memory buffer so that our framework can allocate the tuning budget judiciously in a non-stationary environment. Our framework defines the interfaces adapted to a broad class of ML-based tuning agents, yet simple enough for integration with existing implementations and future extensions. Based on extensive evaluations, we show that this framework can effectively utilize different ML-based agents and find better configurations with 1.4~14.1x speedups on the workload execution time compared with baselines.
- 13 June 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 1 (2) , 1-26
Open AccessA Unified and Efficient Coordinating Framework for Autonomous DBMS Tuning
Recently using machine learning (ML) based techniques to optimize the performance of modern database management systems (DBMSs) has attracted intensive interest from both industry and academia. With an objective to tune a specific component of a DBMS (e.g., index selection, knobs tuning), the ML-based tuning agents have shown to be able to find better configurations than experienced database administrators (DBAs). However, one critical yet challenging question remains unexplored -- how to make those ML-based tuning agents work collaboratively. Existing methods do not consider the dependencies among the multiple agents, and the model used by each agent only studies the effect of changing the configurations in a single component. To tune different components for DBMS, a coordinating mechanism is needed to make the multiple agents be cognizant of each other. Also, we need to decide how to allocate the limited tuning budget (e.g., time and resources) among the agents to maximize the performance. Such a decision is difficult to make since the distribution of the reward (i.e., performance improvement) corresponding to each agent is unknown and non-stationary. In this paper, we study the above question and present a unified coordinating framework to efficiently utilize existing ML-based agents. First, we propose a message propagation protocol that specifies the collaboration behaviors for agents and encapsulates the global tuning messages in each agent's model. Second, we combine Thompson Sampling, a well-studied reinforcement learning algorithm with a memory buffer so that our framework can allocate the tuning budget judiciously in a non-stationary environment. Our framework defines the interfaces adapted to a broad class of ML-based tuning agents, yet simple enough for integration with existing implementations and future extensions. Based on extensive evaluations, we show that this framework can effectively utilize different ML-based agents and find better configurations with 1.4~14.1x speedups on the workload execution time compared with baselines.
- Open Access
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1960-1968
Interactive Log Parsing via Light-weight User Feedback
Template mining is one of the foundational tasks to support log analysis, which supports the diagnosis and troubleshooting of large scale Web applications. This paper develops a human-in-the-loop template mining framework to support interactive log analysis, which is highly desirable in real-world diagnosis or troubleshooting of Web applications but yet previous template mining algorithms fail to support it. We formulate three types of light-weight user feedback and based on them we design three atomic human-in-the-loop template mining algorithms. We derive mild conditions under which the outputs of our proposed algorithms are provably correct. We also derive upper bounds on the computational complexity and query complexity of each algorithm. We demonstrate the versatility of our proposed algorithms by combining them to improve the template mining accuracy of five representative algorithms over sixteen widely used benchmark datasets.
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1960-1968
Open AccessInteractive Log Parsing via Light-weight User Feedback
Template mining is one of the foundational tasks to support log analysis, which supports the diagnosis and troubleshooting of large scale Web applications. This paper develops a human-in-the-loop template mining framework to support interactive log analysis, which is highly desirable in real-world diagnosis or troubleshooting of Web applications but yet previous template mining algorithms fail to support it. We formulate three types of light-weight user feedback and based on them we design three atomic human-in-the-loop template mining algorithms. We derive mild conditions under which the outputs of our proposed algorithms are provably correct. We also derive upper bounds on the computational complexity and query complexity of each algorithm. We demonstrate the versatility of our proposed algorithms by combining them to improve the template mining accuracy of five representative algorithms over sixteen widely used benchmark datasets.
- Open Access
- 25 March 2023
- Published by Association for Computing Machinery (ACM)
- p. 287-323
ShapleyIQ: Influence Quantification by Shapley Values for Performance Debugging of Microservices
Years of experience in operating large-scale microservice systems strengthens our belief that their individual components, with inevitable anomalies, still demand a quantification of the influences on the end-to-end performance indicators. On a causal graph that represents the complex dependencies between the system components, the scatteredly detected anomalies, even when they look similar, could have different implications with contrastive remedial actions. To this end, we design ShapleyIQ, an online monitoring and diagnosis service that can effectively improve the system stability. It is guided by rigorous analysis on Shapley values for a causal graph. Notably, a new property on splitting invariance addresses the challenging exponential computation complexity problem of generic Shapley values by decomposition. This service has been deployed on a core infrastructure system on Alibaba Cloud, for over a year with more than 15,000 operations for 86 services across 2,546 machines. Since then, it has drastically improved the DevOps efficiency, and the system failures have been significantly reduced by 83.3%. We also conduct an offline evaluation on an open source microservice system TrainTicket, which is to pinpoint the root causes of the performance issues in hindsight. Extensive experiments and test cases show that our system can achieve 97.3% accuracy in identifying the top-1 root causes for these datasets, which significantly outperforms baseline algorithms by at least 28.7% in absolute difference.
- 25 March 2023
- Published by Association for Computing Machinery (ACM)
- p. 287-323
Open AccessShapleyIQ: Influence Quantification by Shapley Values for Performance Debugging of Microservices
Years of experience in operating large-scale microservice systems strengthens our belief that their individual components, with inevitable anomalies, still demand a quantification of the influences on the end-to-end performance indicators. On a causal graph that represents the complex dependencies between the system components, the scatteredly detected anomalies, even when they look similar, could have different implications with contrastive remedial actions. To this end, we design ShapleyIQ, an online monitoring and diagnosis service that can effectively improve the system stability. It is guided by rigorous analysis on Shapley values for a causal graph. Notably, a new property on splitting invariance addresses the challenging exponential computation complexity problem of generic Shapley values by decomposition. This service has been deployed on a core infrastructure system on Alibaba Cloud, for over a year with more than 15,000 operations for 86 services across 2,546 machines. Since then, it has drastically improved the DevOps efficiency, and the system failures have been significantly reduced by 83.3%. We also conduct an offline evaluation on an open source microservice system TrainTicket, which is to pinpoint the root causes of the performance issues in hindsight. Extensive experiments and test cases show that our system can achieve 97.3% accuracy in identifying the top-1 root causes for these datasets, which significantly outperforms baseline algorithms by at least 28.7% in absolute difference.
- Open Access
- 5 March 2021
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 48 (3) , 77-78
Prefetching and Caching for Minimizing Service Costs
In practice, prefetching data strategically has been used to improve caching performance. The idea is that data items can either be cached upon request (traditional approach) or prefetched into the cache before the requests actually occur. The caching and prefetching operations compete for the limited cache space, whose size is typically much smaller than the number of data items. A key challenge is to design an optimal prefetching and caching policy, assuming that the future requests can be predicted to a certain extent. This is a non-trivial challenge even under the idealized assumption that future requests are precisely known.
- 5 March 2021
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 48 (3) , 77-78
Open AccessPrefetching and Caching for Minimizing Service Costs
In practice, prefetching data strategically has been used to improve caching performance. The idea is that data items can either be cached upon request (traditional approach) or prefetched into the cache before the requests actually occur. The caching and prefetching operations compete for the limited cache space, whose size is typically much smaller than the number of data items. A key challenge is to design an optimal prefetching and caching policy, assuming that the future requests can be predicted to a certain extent. This is a non-trivial challenge even under the idealized assumption that future requests are precisely known.
- 1 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 10 (21596182) , 9-16
Improving Multi-job MapReduce Scheduling in an Opportunistic Environment
As a state-of-the-art programming model for big data analytics, MapReduce is well suited for parallel processing of large data sets in opportunistic environments. Existing research on MapReduce in opportunistic environment has focused on improving single job performance, the issue of fairness that is critical in the more dominant scenario of multiple concurrent jobs remains unexplored. We address this problem by proposing an opportunistic fair scheduling algorithm, which extends the broadly adopted Fair Scheduler to an environment where nodes are intermittently available with possibly different availability patterns. The proposed scheduler maintains statistics specific to the opportunistic environment, e.g., node availability rates and pairwise availability correlations, and utilizes this information in scheduling decisions to improve fairness. Using a Hadoop-based implementation, we compare our scheduler with the current Hadoop Fair Scheduler on representative benchmarks. Our experiments verify that our scheduler can significantly reduce the variability in job completion times.
- 1 June 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 10 (21596182) , 9-16
Improving Multi-job MapReduce Scheduling in an Opportunistic Environment
As a state-of-the-art programming model for big data analytics, MapReduce is well suited for parallel processing of large data sets in opportunistic environments. Existing research on MapReduce in opportunistic environment has focused on improving single job performance, the issue of fairness that is critical in the more dominant scenario of multiple concurrent jobs remains unexplored. We address this problem by proposing an opportunistic fair scheduling algorithm, which extends the broadly adopted Fair Scheduler to an environment where nodes are intermittently available with possibly different availability patterns. The proposed scheduler maintains statistics specific to the opportunistic environment, e.g., node availability rates and pairwise availability correlations, and utilizes this information in scheduling decisions to improve fairness. Using a Hadoop-based implementation, we compare our scheduler with the current Hadoop Fair Scheduler on representative benchmarks. Our experiments verify that our scheduler can significantly reduce the variability in job completion times.