Weirong Jiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Weirong Jiang has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
Viktor K. Prasanna
University of Southern California, Los Angeles
20 shared publicationsViktor Prasanna
5 shared publications- Viktor Prasanna
University of Southern California
5 shared publications Zhiming Zhang
4 shared publicationsHoang Le
3 shared publicationsYaxuan Qi
Tsinghua National Lab for Information Science and Technology (TNList), Beijing, China
3 shared publications
Latest Publications
- Open Access
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 52-54
Poster: High-Performance Centralized Parallel Data-Plane Verification for Hyper-Scale DCNs
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 52-54
Open AccessPoster: High-Performance Centralized Parallel Data-Plane Verification for Hyper-Scale DCNs
- Open Access
- 18 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 213-220
Automatic Configuration Repair
Networks are error-prone due to misconfigurations, and it is hard to identify the root causes in the configuration and find a repair due to the size and complexity of networks running distributed routing protocols. Thus, we advocate Automatic Configuration Repair (ACR) to reduce the manual effort. Specifically, we draw some insights from the field of Automatic Software Repair (ASR), crystallize some lessons learned from the real-world repair experience of a large service provider, and propose some directions to realize ACR. Inspired by the generate-and-validate approach from ASR, we propose localize-fix-validate as a possible approach to realize ACR.
- 18 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 213-220
Open AccessAutomatic Configuration Repair
Networks are error-prone due to misconfigurations, and it is hard to identify the root causes in the configuration and find a repair due to the size and complexity of networks running distributed routing protocols. Thus, we advocate Automatic Configuration Repair (ACR) to reduce the manual effort. Specifically, we draw some insights from the field of Automatic Software Repair (ASR), crystallize some lessons learned from the real-world repair experience of a large service provider, and propose some directions to realize ACR. Inspired by the generate-and-validate approach from ASR, we propose localize-fix-validate as a possible approach to realize ACR.
- Open Access
- 3 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 81-87
Scaling Data Plane Verification via Parallelization
The data plane verification of networks in hyperscale environments is challenging due to the complexity and size of modern networks. In this paper, we introduce Medusa, a novel verifier that efficiently analyzes large data plane models using parallel processing on multi-core CPUs. First, we propose a new data structure called RANGESET, which overcomes the parallelism limitations of existing popular data structures such as Binary Decision Diagrams (BDD) used in data plane verifiers. Next, we leverage multi-core processing by dividing the network into distinct groups and assigning each group to a separate thread for computation. The results are then integrated for comprehensive verification. By optimizing the use of multi-core systems, we enhance computational efficiency and accelerate the verification process. Experimental results demonstrate that Medusa outperforms existing tools in terms of speed and memory. For instance, in a network with O(10K) devices and O(1M) forwarding rules, Medusa can detect loops in approximately 5 seconds, outperforming other Data Plane Verifiers (DPVs) where some cannot model and analyze the network. Moreover, in networks that we could compare with other state-of-the-art DPVs, Medusa provides a substantial improvement, with speedups up to 600X, 4000X, and 800X compared to alternatives like Flash, APKeep, and Tulkun, respectively.
- 3 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 81-87
Open AccessScaling Data Plane Verification via Parallelization
The data plane verification of networks in hyperscale environments is challenging due to the complexity and size of modern networks. In this paper, we introduce Medusa, a novel verifier that efficiently analyzes large data plane models using parallel processing on multi-core CPUs. First, we propose a new data structure called RANGESET, which overcomes the parallelism limitations of existing popular data structures such as Binary Decision Diagrams (BDD) used in data plane verifiers. Next, we leverage multi-core processing by dividing the network into distinct groups and assigning each group to a separate thread for computation. The results are then integrated for comprehensive verification. By optimizing the use of multi-core systems, we enhance computational efficiency and accelerate the verification process. Experimental results demonstrate that Medusa outperforms existing tools in terms of speed and memory. For instance, in a network with O(10K) devices and O(1M) forwarding rules, Medusa can detect loops in approximately 5 seconds, outperforming other Data Plane Verifiers (DPVs) where some cannot model and analyze the network. Moreover, in networks that we could compare with other state-of-the-art DPVs, Medusa provides a substantial improvement, with speedups up to 600X, 4000X, and 800X compared to alternatives like Flash, APKeep, and Tulkun, respectively.
- 20 August 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 62 (11) , 2169-2182
Data Structure Optimization for Power- Efficient IP Lookup Architectures
Power consumption has become a limiting factor in designing next generation network routers. Recent observation shows that IP lookup engines dominate the power consumption of core routers. Previous work on reducing power consumption of routers mainly focused on network- and system-level optimizations. This paper represents the first thorough study on the data structure optimization for lowering the power consumption in static random access memory (SRAM)-based IP lookup engines. Three different SRAM-based IP lookup architectures are discussed: nonpipelined, simple pipelined, and memory-balanced pipelined architectures. For each architecture, we formulate the problem of power minimization by revisiting the time-space tradeoff in multibit tries. Two distinct multibit trie algorithms are investigated: the expanded trie and the tree bitmap trie, which are widely used in SRAM-based IP lookup solutions. A theoretical framework is proposed to determine the optimal strides for building a multibit trie so that the worst-case power consumption of the IP lookup architecture is minimized. Experiments using real-life routing tables including both IPv4 and IPv6 data sets demonstrate that careful selection of strides in building the multibit tries can reduce the power consumption dramatically. We believe our methodology can be applied to other variants of multibit tries and can help in designing more power-efficient SRAM-based IP lookup architectures.
- 20 August 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 62 (11) , 2169-2182
Data Structure Optimization for Power- Efficient IP Lookup Architectures
Power consumption has become a limiting factor in designing next generation network routers. Recent observation shows that IP lookup engines dominate the power consumption of core routers. Previous work on reducing power consumption of routers mainly focused on network- and system-level optimizations. This paper represents the first thorough study on the data structure optimization for lowering the power consumption in static random access memory (SRAM)-based IP lookup engines. Three different SRAM-based IP lookup architectures are discussed: nonpipelined, simple pipelined, and memory-balanced pipelined architectures. For each architecture, we formulate the problem of power minimization by revisiting the time-space tradeoff in multibit tries. Two distinct multibit trie algorithms are investigated: the expanded trie and the tree bitmap trie, which are widely used in SRAM-based IP lookup solutions. A theoretical framework is proposed to determine the optimal strides for building a multibit trie so that the worst-case power consumption of the IP lookup architecture is minimized. Experiments using real-life routing tables including both IPv4 and IPv6 data sets demonstrate that careful selection of strides in building the multibit tries can reduce the power consumption dramatically. We believe our methodology can be applied to other variants of multibit tries and can help in designing more power-efficient SRAM-based IP lookup architectures.
- 18 August 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 20 (9) , 1668-1680
Scalable Packet Classification on FPGA
Multi-field packet classification has evolved from traditional fixed 5-tuple matching to flexible matching with arbitrary combination of numerous packet header fields. For example, the recently proposed OpenFlow switching requires classifying each packet using up to 12-tuple packet header fields. It has become a great challenge to develop scalable solutions for next-generation packet classification that support higher throughput, larger rule sets and more packet header fields. This paper exploits the abundant parallelism and other desirable features provided by current field-programmable gate arrays (FPGAs), and proposes a decision-tree-based, 2-D multi-pipeline architecture for next-generation packet classification. We revisit the techniques for traditional 5-tuple packet classification and propose several optimization techniques for the state-of-the-art decision-tree-based algorithm. Given a set of 12-tuple rules, we develop a framework to partition the rule set into multiple subsets each of which is built into an optimized decision tree. A tree-to-pipeline mapping scheme is carefully designed to maximize the memory utilization while sustaining high throughput. The implementation results show that our architecture can store either 10K real-life 5-tuple rules or 1K synthetic 12-tuple rules in on-chip memory of a single state-of-the-art FPGA, and sustain 80 and 40 Gbps throughput for minimum size (40 bytes) packets, respectively.
- 18 August 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 20 (9) , 1668-1680
Scalable Packet Classification on FPGA
Multi-field packet classification has evolved from traditional fixed 5-tuple matching to flexible matching with arbitrary combination of numerous packet header fields. For example, the recently proposed OpenFlow switching requires classifying each packet using up to 12-tuple packet header fields. It has become a great challenge to develop scalable solutions for next-generation packet classification that support higher throughput, larger rule sets and more packet header fields. This paper exploits the abundant parallelism and other desirable features provided by current field-programmable gate arrays (FPGAs), and proposes a decision-tree-based, 2-D multi-pipeline architecture for next-generation packet classification. We revisit the techniques for traditional 5-tuple packet classification and propose several optimization techniques for the state-of-the-art decision-tree-based algorithm. Given a set of 12-tuple rules, we develop a framework to partition the rule set into multiple subsets each of which is built into an optimized decision tree. A tree-to-pipeline mapping scheme is carefully designed to maximize the memory utilization while sustaining high throughput. The implementation results show that our architecture can store either 10K real-life 5-tuple rules or 1K synthetic 12-tuple rules in on-chip memory of a single state-of-the-art FPGA, and sustain 80 and 40 Gbps throughput for minimum size (40 bytes) packets, respectively.
- Open Access
Bidirectional Pipelining for Scalable IP Lookup and Packet Classification
Both IP lookup and packet classification in IP routers can be implemented by some form of tree traversal. SRAM-based Pipelining can improve the throughput dramatically. However, previous pipelining schemes result in unbalanced memory allocation over the pipeline stages. This has been identified as a major challenge for scalable pipelined solutions. This paper proposes a flexible bidirectional linear pipeline architecture based on widely-used dual-port SRAMs. A search tree is partitioned, and then mapped onto pipeline stages by a bidirectional fine-grained mapping scheme. We introduce the notion of inversion factor and several heuristics to invert subtrees for memory balancing. Due to its linear structure, the architecture maintains packet input order, and supports non-blocking route updates. Our experiments show that, the architecture can achieve a perfectly balanced memory distribution over the pipeline stages, for both trie-based IP lookup and tree-based multi-dimensional packet classification. For IP lookup, it can store a full backbone routing table with 154419 entries using 2MB of memory, and sustain a high throughput of 1.87 billion packets per second (GPPS), i.e. 0.6 Tbps for the minimum size (40 bytes) packets. The throughput can be improved further to be 2.4 Tbps, by employing caching to exploit the Internet traffic locality.
Open AccessBidirectional Pipelining for Scalable IP Lookup and Packet Classification
Both IP lookup and packet classification in IP routers can be implemented by some form of tree traversal. SRAM-based Pipelining can improve the throughput dramatically. However, previous pipelining schemes result in unbalanced memory allocation over the pipeline stages. This has been identified as a major challenge for scalable pipelined solutions. This paper proposes a flexible bidirectional linear pipeline architecture based on widely-used dual-port SRAMs. A search tree is partitioned, and then mapped onto pipeline stages by a bidirectional fine-grained mapping scheme. We introduce the notion of inversion factor and several heuristics to invert subtrees for memory balancing. Due to its linear structure, the architecture maintains packet input order, and supports non-blocking route updates. Our experiments show that, the architecture can achieve a perfectly balanced memory distribution over the pipeline stages, for both trie-based IP lookup and tree-based multi-dimensional packet classification. For IP lookup, it can store a full backbone routing table with 154419 entries using 2MB of memory, and sustain a high throughput of 1.87 billion packets per second (GPPS), i.e. 0.6 Tbps for the minimum size (40 bytes) packets. The throughput can be improved further to be 2.4 Tbps, by employing caching to exploit the Internet traffic locality.
- Open Access
- 1 June 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Multiroot: Towards Memory-Efficient Router Virtualization
Network virtualization has become a powerful scheme to make efficient use of networking hardware. It allows multiple virtual networks to co-exist on the same physical networking substrate. This requires the hardware router to maintain multiple lookup tables. Hence, ultimately the hardware router should be capable of handling packets from different virtual networks. In this paper, we introduce a memory-efficient solution for router virtualization named, Multiroot. We propose this potential scheme for Provider Edge (PE) router virtualization after examining the address space requirement of such networks. Multiroot is a novel merging technique to consolidate all the routing tables to a single merged table. The shared data structure used in our algorithm results in a significant memory usage reduction in the lookup data structure while guaranteeing traffic isolation which is critical in a virtualized environment. This improvement in memory usage results in a very scalable solution for router virtualization in terms of resource usage of the hardware router. Multiroot uses trie data structure and can be implemented on a hardware or a software platform. Experiments show that our solution can achieve up to 5 fold memory usage reduction compared to state-of-the-art techniques present in literature.
- 1 June 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Open AccessMultiroot: Towards Memory-Efficient Router Virtualization
Network virtualization has become a powerful scheme to make efficient use of networking hardware. It allows multiple virtual networks to co-exist on the same physical networking substrate. This requires the hardware router to maintain multiple lookup tables. Hence, ultimately the hardware router should be capable of handling packets from different virtual networks. In this paper, we introduce a memory-efficient solution for router virtualization named, Multiroot. We propose this potential scheme for Provider Edge (PE) router virtualization after examining the address space requirement of such networks. Multiroot is a novel merging technique to consolidate all the routing tables to a single merged table. The shared data structure used in our algorithm results in a significant memory usage reduction in the lookup data structure while guaranteeing traffic isolation which is critical in a virtualized environment. This improvement in memory usage results in a very scalable solution for router virtualization in terms of resource usage of the hardware router. Multiroot uses trie data structure and can be implemented on a hardware or a software platform. Experiments show that our solution can achieve up to 5 fold memory usage reduction compared to state-of-the-art techniques present in literature.
- 1 May 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 242-249
Memory-Efficient IPv4/v6 Lookup on FPGAs Using Distance-Bounded Path Compression
Memory efficiency with compact data structures for Internet Protocol (IP) lookup has recently regained much interest in the research community. In this paper, we revisit the classic trie-based approach for solving the longest prefix matching (LPM) problem used in IP lookup. In particular, we target our solutions for a class of large and sparsely-distributed routing tables, such as those potentially arising in the next generation IPv6 routing protocol. Due to longer prefix and much larger address space, straight-forward implementation of trie based LPM can significantly increase the number of nodes and/or memory required for IP lookup. Additionally, due to the available on-chip memory and the number of I/O pins of in the state-of-the art Field Programmable Gate Arrays (FPGAs), existing designs cannot support large IPv6 routing tables consisting of over 300K prefixes. We propose two algorithms to compress the uni-bit-trie representation of a given routing table: (1) single-prefix distance bounded path compression and (2) multiple-prefix distance bounded path compression. These algorithms determine the optimal maximum slap distance at each node of the trie to minimize the total memory requirement. Our algorithms demonstrate substantial reduction in the memory footprint compared with the uni-bit-trie algorithm (1.86× for IPv4 and 6.16× for IPv6), and with the original path compression algorithm (1.77× for IPv4 and 1.53× for IPv6). Furthermore, implementation on a state of-the-art FPGA shows that our algorithms achieve 466 million lookups per second and are well suited for 100Gbps lookup. The implementation also scales to support larger routing tables and longer prefix length when we go from IPv4 to IPv6.
- 1 May 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 242-249
Memory-Efficient IPv4/v6 Lookup on FPGAs Using Distance-Bounded Path Compression
Memory efficiency with compact data structures for Internet Protocol (IP) lookup has recently regained much interest in the research community. In this paper, we revisit the classic trie-based approach for solving the longest prefix matching (LPM) problem used in IP lookup. In particular, we target our solutions for a class of large and sparsely-distributed routing tables, such as those potentially arising in the next generation IPv6 routing protocol. Due to longer prefix and much larger address space, straight-forward implementation of trie based LPM can significantly increase the number of nodes and/or memory required for IP lookup. Additionally, due to the available on-chip memory and the number of I/O pins of in the state-of-the art Field Programmable Gate Arrays (FPGAs), existing designs cannot support large IPv6 routing tables consisting of over 300K prefixes. We propose two algorithms to compress the uni-bit-trie representation of a given routing table: (1) single-prefix distance bounded path compression and (2) multiple-prefix distance bounded path compression. These algorithms determine the optimal maximum slap distance at each node of the trie to minimize the total memory requirement. Our algorithms demonstrate substantial reduction in the memory footprint compared with the uni-bit-trie algorithm (1.86× for IPv4 and 6.16× for IPv6), and with the original path compression algorithm (1.77× for IPv4 and 1.53× for IPv6). Furthermore, implementation on a state of-the-art FPGA shows that our algorithms achieve 466 million lookups per second and are well suited for 100Gbps lookup. The implementation also scales to support larger routing tables and longer prefix length when we go from IPv4 to IPv6.
- 1 April 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 0743166X,p. 2114-2122
FEACAN: Front-end acceleration for content-aware network processing
Modern networks are increasingly becoming content aware to improve data delivery and security via content-based network processing. Content-aware processing at the front end of distributed network systems, such as application identification for datacenter load-balancers and deep packet inspection for security gateways, is more challenging due to the wire-speed and low-latency requirement. Existing work focuses on algorithm-level solutions while lacking system-level design to meet the critical requirement for front-end content processing. In this paper, we propose a system-level solution named FEACAN for front-end acceleration of content-aware network processing. FEACAN employs a software-hardware co-design supporting both signature matching and regular expression matching for content-aware network processing. A two-dimensional DFA compression algorithm is designed to reduce the memory usage and a hardware lookup engine is proposed for high-performance lookup. Experimental results show that FEACAN achieves better performance than existing work in terms of processing speed, resource utilization, and update time.
- 1 April 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 0743166X,p. 2114-2122
FEACAN: Front-end acceleration for content-aware network processing
Modern networks are increasingly becoming content aware to improve data delivery and security via content-based network processing. Content-aware processing at the front end of distributed network systems, such as application identification for datacenter load-balancers and deep packet inspection for security gateways, is more challenging due to the wire-speed and low-latency requirement. Existing work focuses on algorithm-level solutions while lacking system-level design to meet the critical requirement for front-end content processing. In this paper, we propose a system-level solution named FEACAN for front-end acceleration of content-aware network processing. FEACAN employs a software-hardware co-design supporting both signature matching and regular expression matching for content-aware network processing. A two-dimensional DFA compression algorithm is designed to reduce the memory usage and a hardware lookup engine is proposed for high-performance lookup. Experimental results show that FEACAN achieves better performance than existing work in terms of processing speed, resource utilization, and update time.