Huiqi Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Huiqi Hu has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
Aoying Zhou
School of Data Science and Engineering, East China Normal University, Shanghai, China
19 shared publicationsWeining Qian
School of Data Science and Engineering, East China Normal University, Shanghai, China
12 shared publicationsXuan Zhou
School of Data Science and Engineering, East China Normal University, Shanghai, China
10 shared publications- Xuan Zhou
East China Normal University, Shanghai, China
9 shared publications - Aoying Zhou
East China Normal University, Shanghai, China
5 shared publications Chenchen Huang
School of Data Science and Engineering, East China Normal University, Shanghai, China
5 shared publications
Latest Publications
- Open Access
- 7 October 2026
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Architecture and Code Optimization
Nephele: Bridging DDRāCXL Latency Gaps with Benefit-Aware Caching for In-Memory Databases on Hybrid Memory
In-memory databases rely on DDR memory for performance, yet capacity scaling is impeded by physical constraints and the high cost of high-density DIMMs. Compute Express Link (CXL) provides a practical path for memory expansion through a DDR-compatible interface. However, CXL memory exhibits inferior latency and bandwidth, thus naĆÆve expansion with CXL can lead to significant performance degradation. In hybrid DDRāCXL systems, traditional caching policies (e.g., LRU, S3-FIFO) yield limited reductions in execution latency while incurring substantial metadata management and tuple migration overheads. Our analysis reveals a fundamental misalignment between traditional design assumptions and heterogeneous memory characteristics. To address this, we propose Nephele, a lightweight, benefit-aware caching policy for hybrid DDRāCXL databases. Nephele shifts the focus from eviction selection to effective admission, prioritizing data that are truly worth caching. It also features a low-overhead architecture that supports randomized eviction without relying on global metadata. By integrating epoch-based hotness estimation, Nephele enables probabilistic, benefit-aware admission, reducing cross-tier data movement and mitigating cache pollution. Evaluated on a hybrid DDRāCXL database prototype using YCSB and TPC-C, Nephele achieves up to 1.58Ć higher throughput than CXL-Expansion and up to 3.2Ć over existing caching policies, while retaining up to 97% of DDR-only performance. Additional integration with RocksDB further confirms its practical feasibility in a production system. These results demonstrate that lightweight, benefit-aware caching is critical for fully exploiting CXL-enabled memory expansion.
- 7 October 2026
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Architecture and Code Optimization
Open AccessNephele: Bridging DDRāCXL Latency Gaps with Benefit-Aware Caching for In-Memory Databases on Hybrid Memory
In-memory databases rely on DDR memory for performance, yet capacity scaling is impeded by physical constraints and the high cost of high-density DIMMs. Compute Express Link (CXL) provides a practical path for memory expansion through a DDR-compatible interface. However, CXL memory exhibits inferior latency and bandwidth, thus naĆÆve expansion with CXL can lead to significant performance degradation. In hybrid DDRāCXL systems, traditional caching policies (e.g., LRU, S3-FIFO) yield limited reductions in execution latency while incurring substantial metadata management and tuple migration overheads. Our analysis reveals a fundamental misalignment between traditional design assumptions and heterogeneous memory characteristics. To address this, we propose Nephele, a lightweight, benefit-aware caching policy for hybrid DDRāCXL databases. Nephele shifts the focus from eviction selection to effective admission, prioritizing data that are truly worth caching. It also features a low-overhead architecture that supports randomized eviction without relying on global metadata. By integrating epoch-based hotness estimation, Nephele enables probabilistic, benefit-aware admission, reducing cross-tier data movement and mitigating cache pollution. Evaluated on a hybrid DDRāCXL database prototype using YCSB and TPC-C, Nephele achieves up to 1.58Ć higher throughput than CXL-Expansion and up to 3.2Ć over existing caching policies, while retaining up to 97% of DDR-only performance. Additional integration with RocksDB further confirms its practical feasibility in a production system. These results demonstrate that lightweight, benefit-aware caching is critical for fully exploiting CXL-enabled memory expansion.
- Open Access
- 4 December 2025
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ Proceedings of the ACM on Management of Data
- Vol.Ā 3 Ā (6) , 1-25
APQO: An Adaptive Framework for Parametric Query Optimization
Previous learning-based parameter query optimization (PQO) methods excel in static workloads by precisely selecting optimal plans in a cache with a fixed set of representative plans. However, these methods struggle in dynamic workloads because they cannot predict over dynamically changed plan caches. These queries that fall outside the previously observed query parameter distribution have the risk of reusing suboptimal plans. Unlike traditional PQO methods that learn mappings from parametric query parameters to a fixed set of execution plans, our adaptive PQO framework ( APQO ) takes both query parameters and the plans themselves as model inputs, thereby handling variable numbers of plans in dynamic workloads. By embedding plan representations through representation learning, we pre-train a foundation model offline, enabling APQO to acquire a generalizable plan performance prediction model. Leveraging the foundation model's predictive capability along with a hybrid data augmentation strategy, we train an online calibration model with minimal training data for distribution-shifted new queries, rapidly adapting knowledge for reusing new plans. APQO is natively designed to handle the characteristics of dynamic workloads. Experimental results show that APQO outperforms existing PQO methods in dynamic workloads, achieving a higher cache hit ratio and significantly reducing query latency.
- 4 December 2025
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ Proceedings of the ACM on Management of Data
- Vol.Ā 3 Ā (6) , 1-25
Open AccessAPQO: An Adaptive Framework for Parametric Query Optimization
Previous learning-based parameter query optimization (PQO) methods excel in static workloads by precisely selecting optimal plans in a cache with a fixed set of representative plans. However, these methods struggle in dynamic workloads because they cannot predict over dynamically changed plan caches. These queries that fall outside the previously observed query parameter distribution have the risk of reusing suboptimal plans. Unlike traditional PQO methods that learn mappings from parametric query parameters to a fixed set of execution plans, our adaptive PQO framework ( APQO ) takes both query parameters and the plans themselves as model inputs, thereby handling variable numbers of plans in dynamic workloads. By embedding plan representations through representation learning, we pre-train a foundation model offline, enabling APQO to acquire a generalizable plan performance prediction model. Leveraging the foundation model's predictive capability along with a hybrid data augmentation strategy, we train an online calibration model with minimal training data for distribution-shifted new queries, rapidly adapting knowledge for reusing new plans. APQO is natively designed to handle the characteristics of dynamic workloads. Experimental results show that APQO outperforms existing PQO methods in dynamic workloads, achieving a higher cache hit ratio and significantly reducing query latency.
- 7 August 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Parallel and Distributed Systems
- Vol.Ā 36 Ā (10) , 2137-2151
Mariana: Exploring Native SkipList Index Design for Disaggregated Memory
Memory disaggregation has emerged as a promising architecture for improving resource efficiency by decoupling the computing and memory resources. But building efficient range indices in such an architecture faces three critical challenges: (1) coarse-grained concurrency control schemes for coordinating concurrent read/write operations with node splitting incur high contention under the skewed and write-intensive workloads; (2) existing data layouts fail to balance consistency verification and hardware acceleration via SIMD (Single Instruction Multiple Data); and (3) naive caching schemes struggle to adapt to rapidly changing access patterns. To address these challenges, we propose Mariana, a memory-disaggregated skiplist index that integrates three key innovations. First, it uses a fine-grained (i.e., entry-level) latch mechanism combined with dynamic node resizing to minimize the contention and splitting frequency. Second, it employs a tailored data layout for leaf node, which separates keys and values to enable SIMD acceleration while maintaining consistency checks with minimal write overhead. Third, it implements an adaptive caching strategy that tracks node popularity in real-time to optimize network bandwidth utilization during the index traversal. Experimental results show that Mariana achieves$1.7\times$ higher throughput under write-intensive workloads and reduces the P90 latency by 23% under the read-intensive workloads, when comparing to the state-of-the-art indices on disaggregated memory.- 7 August 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on Parallel and Distributed Systems
- Vol.Ā 36 Ā (10) , 2137-2151
Mariana: Exploring Native SkipList Index Design for Disaggregated Memory
Memory disaggregation has emerged as a promising architecture for improving resource efficiency by decoupling the computing and memory resources. But building efficient range indices in such an architecture faces three critical challenges: (1) coarse-grained concurrency control schemes for coordinating concurrent read/write operations with node splitting incur high contention under the skewed and write-intensive workloads; (2) existing data layouts fail to balance consistency verification and hardware acceleration via SIMD (Single Instruction Multiple Data); and (3) naive caching schemes struggle to adapt to rapidly changing access patterns. To address these challenges, we propose Mariana, a memory-disaggregated skiplist index that integrates three key innovations. First, it uses a fine-grained (i.e., entry-level) latch mechanism combined with dynamic node resizing to minimize the contention and splitting frequency. Second, it employs a tailored data layout for leaf node, which separates keys and values to enable SIMD acceleration while maintaining consistency checks with minimal write overhead. Third, it implements an adaptive caching strategy that tracks node popularity in real-time to optimize network bandwidth utilization during the index traversal. Experimental results show that Mariana achieves$1.7\times$ higher throughput under write-intensive workloads and reduces the P90 latency by 23% under the read-intensive workloads, when comparing to the state-of-the-art indices on disaggregated memory.- Open Access
SynchroStore: A Cost-Based Fine-Grained Incremental Compaction for Hybrid Workloads
This study proposes a novel storage engine, SynchroStore, designed to address the inefficiency of update operations in columnar storage systems based on Log-Structured Merge Trees (LSM-Trees) under hybrid workload scenarios. While columnar storage formats demonstrate significant query performance advantages when handling large-scale datasets, traditional columnar storage systems face challenges such as high update complexity and poor real-time performance in data-intensive applications. SynchroStore introduces an incremental row storage mechanism and a fine-grained row-to-column transformation and compaction strategy, effectively balancing data update efficiency and query performance. The storage system employs an in-memory row storage structure to support efficient update operations, and the data is converted to a columnar format after freezing to support high-performance read operations. The core innovations of SynchroStore are reflected in the following aspects:(1) the organic combination of incremental row storage and columnar storage; (2) a fine-grained row-to-column transformation and compaction mechanism; (3) a cost-based scheduling strategy. These innovative features allow SynchroStore to leverage background computational resources for row-to-column transformation and compaction operations, while ensuring query performance is unaffected, thus effectively solving the update performance bottleneck of columnar storage under hybrid workloads. Experimental evaluation results show that, compared to existing columnar storage systems like DuckDB, SynchroStore exhibits significant advantages in update performance under hybrid workloads.
Open AccessSynchroStore: A Cost-Based Fine-Grained Incremental Compaction for Hybrid Workloads
This study proposes a novel storage engine, SynchroStore, designed to address the inefficiency of update operations in columnar storage systems based on Log-Structured Merge Trees (LSM-Trees) under hybrid workload scenarios. While columnar storage formats demonstrate significant query performance advantages when handling large-scale datasets, traditional columnar storage systems face challenges such as high update complexity and poor real-time performance in data-intensive applications. SynchroStore introduces an incremental row storage mechanism and a fine-grained row-to-column transformation and compaction strategy, effectively balancing data update efficiency and query performance. The storage system employs an in-memory row storage structure to support efficient update operations, and the data is converted to a columnar format after freezing to support high-performance read operations. The core innovations of SynchroStore are reflected in the following aspects:(1) the organic combination of incremental row storage and columnar storage; (2) a fine-grained row-to-column transformation and compaction mechanism; (3) a cost-based scheduling strategy. These innovative features allow SynchroStore to leverage background computational resources for row-to-column transformation and compaction operations, while ensuring query performance is unaffected, thus effectively solving the update performance bottleneck of columnar storage under hybrid workloads. Experimental evaluation results show that, compared to existing columnar storage systems like DuckDB, SynchroStore exhibits significant advantages in update performance under hybrid workloads.
- Open Access
- 15 January 2025
- Published byĀ Springer NatureĀ inĀ Data Science and Engineering
- Vol.Ā 10 Ā (2) , 212-229
Aion: Live Migration for In-Memory Databases with Zero Downtime and Reduced Redundant Data Transfer
Distributed in-memory databases are widely adopted to achieve low latency and high bandwidth for data-intensive applications. They support scale-out by sharding and distributing data across multiple nodes. To efficiently adapt to various workloads, distributed in-memory databases must be capable of migrating shards across nodes. In this paper, we demonstrate that state-of-the-art approaches experience significant performance degradation during migration due to service downtime and redundant data transfer. Furthermore, our findings indicate that the presence of service downtime constrains the scalability of migration strategies, while the transfer of redundant data during the snapshot transfer phase limits their adaptability to dynamic workloads. To this end, this paper proposes Aion, a live migration strategy designed for distributed in-memory databases. Aion eliminates any potential service downtime by immediately switching transaction routing to the destination node. To ensure data consistency between the source and destination nodes, as well as serializable execution during migration, Aion proposes the mutual validation phase. Moreover, Aion introduces an analysis phase before the snapshot transfer phase to identify dynamically changing hotspots in workloads. The analysis phase identifies and transfers tuples and versions accessed less frequently to the destination node, reducing the amount of data transferred. Aion is implemented on a distributed in-memory database and evaluated using various OLTP workloads. The results demonstrate that Aion can fundamentally eliminate service downtime, adapt effectively to various workloads and exhibit robust scalability. Compared to state-of-the-art approaches, Aion achieves up to 2.25xā6.57x higher throughput during migration and shortens the migration duration by 53.7ā68.2%.
- 15 January 2025
- Published byĀ Springer NatureĀ inĀ Data Science and Engineering
- Vol.Ā 10 Ā (2) , 212-229
Open AccessAion: Live Migration for In-Memory Databases with Zero Downtime and Reduced Redundant Data Transfer
Distributed in-memory databases are widely adopted to achieve low latency and high bandwidth for data-intensive applications. They support scale-out by sharding and distributing data across multiple nodes. To efficiently adapt to various workloads, distributed in-memory databases must be capable of migrating shards across nodes. In this paper, we demonstrate that state-of-the-art approaches experience significant performance degradation during migration due to service downtime and redundant data transfer. Furthermore, our findings indicate that the presence of service downtime constrains the scalability of migration strategies, while the transfer of redundant data during the snapshot transfer phase limits their adaptability to dynamic workloads. To this end, this paper proposes Aion, a live migration strategy designed for distributed in-memory databases. Aion eliminates any potential service downtime by immediately switching transaction routing to the destination node. To ensure data consistency between the source and destination nodes, as well as serializable execution during migration, Aion proposes the mutual validation phase. Moreover, Aion introduces an analysis phase before the snapshot transfer phase to identify dynamically changing hotspots in workloads. The analysis phase identifies and transfers tuples and versions accessed less frequently to the destination node, reducing the amount of data transferred. Aion is implemented on a distributed in-memory database and evaluated using various OLTP workloads. The results demonstrate that Aion can fundamentally eliminate service downtime, adapt effectively to various workloads and exhibit robust scalability. Compared to state-of-the-art approaches, Aion achieves up to 2.25xā6.57x higher throughput during migration and shortens the migration duration by 53.7ā68.2%.
- Open Access
- 1 January 2025
- Published byĀ International Journal of Software and Informatics (IJSI)Ā inĀ International Journal of Software and Informatics
- Vol.Ā 15 Ā (1) , 43-67
PG-RAC: PostgreSQL-based Database with Shared Cache for Multi-write Transaction
Single-master is the mainstream architecture of cloud-native databases. In the cluster, the slave node can share the read-only requests of the master node, and the master node processes the write requests. To meet the demands of large-scale transaction processing, on this basis, some of the cloud databases attempt to further extend multi-write transactions. One way to implement a multi-write extension is to implement a shared cache among compute nodes and support cross-node data access. In database systems based on shared cache, the cost of cross-node remote access is much larger than that of local access, so the design of the cache protocol is the key factor affecting the system's performance and scalability. In this study, two innovative improvements to the cache protocol are proposed, and PG-RAC, a shared cache database supporting multi-write transaction processing, is implemented based on PostgreSQL. On the one hand, PG-RAC proposes a new distributed chained routing strategy to distribute routing information among compute nodes. The average transaction latency is reduced by about 20% compared to the routing strategy of single-point directory management. On the other hand, the replica invalidation mechanism is improved, and the invalidation operation is separated from the transaction path, reducing the latency of the critical path in transaction processing. On this basis, PG-RAC utilizes the characteristics of multi-version concurrency control (MVCC) to further propose to delay replica invalidation, effectively improving the cache utilization. The results of the TPC-C experiment show that in a cluster with four compute nodes, the throughput rate is nearly twice that of PostgreSQL and 1.5 times that of the distributed database Citus.
- 1 January 2025
- Published byĀ International Journal of Software and Informatics (IJSI)Ā inĀ International Journal of Software and Informatics
- Vol.Ā 15 Ā (1) , 43-67
Open AccessPG-RAC: PostgreSQL-based Database with Shared Cache for Multi-write Transaction
Single-master is the mainstream architecture of cloud-native databases. In the cluster, the slave node can share the read-only requests of the master node, and the master node processes the write requests. To meet the demands of large-scale transaction processing, on this basis, some of the cloud databases attempt to further extend multi-write transactions. One way to implement a multi-write extension is to implement a shared cache among compute nodes and support cross-node data access. In database systems based on shared cache, the cost of cross-node remote access is much larger than that of local access, so the design of the cache protocol is the key factor affecting the system's performance and scalability. In this study, two innovative improvements to the cache protocol are proposed, and PG-RAC, a shared cache database supporting multi-write transaction processing, is implemented based on PostgreSQL. On the one hand, PG-RAC proposes a new distributed chained routing strategy to distribute routing information among compute nodes. The average transaction latency is reduced by about 20% compared to the routing strategy of single-point directory management. On the other hand, the replica invalidation mechanism is improved, and the invalidation operation is separated from the transaction path, reducing the latency of the critical path in transaction processing. On this basis, PG-RAC utilizes the characteristics of multi-version concurrency control (MVCC) to further propose to delay replica invalidation, effectively improving the cache utilization. The results of the TPC-C experiment show that in a cluster with four compute nodes, the throughput rate is nearly twice that of PostgreSQL and 1.5 times that of the distributed database Citus.
- Open Access
- 20 November 2024
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Architecture and Code Optimization
- Vol.Ā 21 Ā (4) , 1-26
SuccinctKV: a CPU-efficient LSM-tree Based KV Store with Scan-based Compaction
The CPU overhead of the LSM-tree becomes increasingly significant when high-speed storage devices are utilized. In this article, we propose SuccinctKV , a key-value store based on LSM-tree that is optimized to improve CPU efficiency in mixed workload scenarios. To achieve this, SuccinctKV reduces the CPU overhead of compaction by writing scan-sorted data directly to the storage device. SuccinctKV also redesigns the merge-sort operation of the LSM-tree, enhancing CPU locality and reducing the unnecessary CPU overhead of cache accesses and I/O system calls. Additionally, SuccinctKV introduces a scheduler to resolve potential bursty I/O contention by autonomously initiating I/O requests at appropriate times and quickly relieving I/O pressure by terminating background I/O requests. We implement SuccinctKV on RocksDB and conduct extensive experiments to evaluate our proposed methods. The experimental results demonstrate that, compared with RocksDB, SuccinctKV achieves a maximum improvement of 2.6x in scan performance and reduces CPU overhead of compaction by up to 89% under mixed workloads.
- 20 November 2024
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Architecture and Code Optimization
- Vol.Ā 21 Ā (4) , 1-26
Open AccessSuccinctKV: a CPU-efficient LSM-tree Based KV Store with Scan-based Compaction
The CPU overhead of the LSM-tree becomes increasingly significant when high-speed storage devices are utilized. In this article, we propose SuccinctKV , a key-value store based on LSM-tree that is optimized to improve CPU efficiency in mixed workload scenarios. To achieve this, SuccinctKV reduces the CPU overhead of compaction by writing scan-sorted data directly to the storage device. SuccinctKV also redesigns the merge-sort operation of the LSM-tree, enhancing CPU locality and reducing the unnecessary CPU overhead of cache accesses and I/O system calls. Additionally, SuccinctKV introduces a scheduler to resolve potential bursty I/O contention by autonomously initiating I/O requests at appropriate times and quickly relieving I/O pressure by terminating background I/O requests. We implement SuccinctKV on RocksDB and conduct extensive experiments to evaluate our proposed methods. The experimental results demonstrate that, compared with RocksDB, SuccinctKV achieves a maximum improvement of 2.6x in scan performance and reduces CPU overhead of compaction by up to 89% under mixed workloads.
- 1 October 2024
- Published byĀ Springer Nature
- p. 315-330
SPQO: Learning to Safely Reuse Cached Plans for Dynamic Workloads
- 1 October 2024
- Published byĀ Springer Nature
- p. 315-330
SPQO: Learning to Safely Reuse Cached Plans for Dynamic Workloads
- 28 August 2024
- Published byĀ Springer Nature
- p. 428-443
A Composable Architecture for Cloud Transactional DBMS
- 28 August 2024
- Published byĀ Springer Nature
- p. 428-443
A Composable Architecture for Cloud Transactional DBMS
- Open Access
Fast Commitment for Geo-Distributed Transactions via Decentralized Co-coordinators
In a geo-distributed database, data shards and their respective replicas are deployed in distinct datacenters across multiple regions, enabling regional-level disaster recovery and the ability to serve global users locally. However, transaction processing in geo-distributed databases requires multiple cross-region communications, especially during the commit phase, which can significantly impact system performance. To optimize the performance of geo-distributed transactions, we propose Decentralized Two-phase Commit (D2PC), a new transaction commit protocol aiming to minimize the negative impact of cross-region communication. In D2PC, we employ multiple co-coordinators that perform commit coordination in parallel. Each co-coordinator is responsible for collecting 2PC votes and making a PreCommit decision in its local region. This approach allows for the concurrent invocation of multiple cross-region network round trips, and each region can conclude its concurrency control locally before replication is complete, thus significantly reducing the chances of blocking and enhancing system concurrency. Moreover, we propose the bypass leader replication reply method, leveraging decentralized co-coordinators to bypass the leader for message transmission, thereby reducing the commit latency. Experimental results have demonstrated that D2PC can reduce commit latency by 43% and improve throughput by up to 2.43 times compared to the existing alternative geo-distributed transaction processing methods.
Open AccessFast Commitment for Geo-Distributed Transactions via Decentralized Co-coordinators
In a geo-distributed database, data shards and their respective replicas are deployed in distinct datacenters across multiple regions, enabling regional-level disaster recovery and the ability to serve global users locally. However, transaction processing in geo-distributed databases requires multiple cross-region communications, especially during the commit phase, which can significantly impact system performance. To optimize the performance of geo-distributed transactions, we propose Decentralized Two-phase Commit (D2PC), a new transaction commit protocol aiming to minimize the negative impact of cross-region communication. In D2PC, we employ multiple co-coordinators that perform commit coordination in parallel. Each co-coordinator is responsible for collecting 2PC votes and making a PreCommit decision in its local region. This approach allows for the concurrent invocation of multiple cross-region network round trips, and each region can conclude its concurrency control locally before replication is complete, thus significantly reducing the chances of blocking and enhancing system concurrency. Moreover, we propose the bypass leader replication reply method, leveraging decentralized co-coordinators to bypass the leader for message transmission, thereby reducing the commit latency. Experimental results have demonstrated that D2PC can reduce commit latency by 43% and improve throughput by up to 2.43 times compared to the existing alternative geo-distributed transaction processing methods.