The Experts below are selected from a list of 309 Experts worldwide ranked by ideXlab platform
Patrick Crowley - One of the best experts on this subject based on the ideXlab platform.
-
Data structures, algorithms and architectures for efficient regular Expression Evaluation
2020Co-Authors: Patrick Crowley, Michela BecchiAbstract:Regular Expression matching is a crucial task in several networking applications, where packet payloads need to be inspected at line rate over large sets of complex patterns. Modern, payload-scanning firewalls and intrusion detection systems are perhaps the most notable applications that rely on this mechanism. When targeting memory-based architectures, in which the pattern data structure is stored in off-chip memory such as DRAM, the design of efficient regular Expression matching engines is subject to two basic resource requirements: memory size and memory bandwidth. When designing Field Programmable Gate Array implementations, in which the pattern data structure is stored in on-chip memories or logic gates, the focus is on minimizing logic cell utilization and allowing high operational dock frequency. Keeping these requirements feasible is a challenge as pattern sets grow in size and expressiveness. In this work, we describe novel data structures and algorithms for high-speed regular Expression Evaluation and map the proposed schemes onto suitable parallel architectures. We aim to provide a comprehensive solution that is scalable with respect to both the size of the rule-set and the complexity of the individual rules. Our main contributions are the following. First, we propose an efficient and low complexity compression scheme for deterministic finite automata (DFAs), leading to a compression factor in excess of 98% across a set of representative data sets. Second, we identify and address problematic subpatterns, such as bounded and unbounded repetitions of wildcards and large character sets and back-references. Specifically, we propose a novel automaton with a limited memory footprint and a bounded memory bandwidth requirement that overcomes the limitations of pure DFA-based solutions. Third, we describe how to utilize our proposed techniques when targeting FPGA implementations. Finally, our tools for constructing and analyzing a range of alternative automata for regular-Expression Evaluation have been released as open-source, and are used actively by many researchers from around the world.
-
a dfa a time and space efficient dfa compression algorithm for fast regular Expression Evaluation
ACM Transactions on Architecture and Code Optimization, 2013Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Modern network intrusion detection systems need to perform regular Expression matching at line rate in order to detect the occurrence of critical patterns in packet payloads. While Deterministic Finite Automata (DFAs) allow this operation to be performed in linear time, they may exhibit prohibitive memory requirements. Kumar et al. [2006a] have proposed Delayed Input DFAs (D2FAs), which provide a trade-off between the memory requirements of the compressed DFA and the number of states visited for each character processed, which in turn affects the memory bandwidth required to evaluate regular Expressions. In this article we introduce Amortized time − bandwidth overhead DFAs (A − DFAs), a general compression technique that results in at most N(k + 1)/k state traversals when processing a string of length N, k being a positive integer. In comparison to the D2FA approach, our technique achieves comparable levels of compression with lower provable bounds on memory bandwidth (or greater compression for a given bandwidth bound). Moreover, the A-DFA algorithm has lower complexity, can be applied during DFA creation, and is suitable for scenarios where a compressed DFA needs to be dynamically built or updated. Finally, we show how to combine A-DFA with alphabet reduction and multistride DFAs, two techniques aimed at reducing the memory space and bandwidth requirement of DFAs, and discuss memory encoding schemes suitable for A-DFAs.
-
ANCS - Efficient regular Expression Evaluation: theory to practice
Proceedings of the 4th ACM IEEE Symposium on Architectures for Networking and Communications Systems - ANCS '08, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been proposed recently to accelerate regular Expression matching and enable deep packet inspection at line rate. This work aims to provide a comprehensive practical Evaluation of existing techniques, extending them and analyzing their compatibility. The study focuses on two hardware architectures: memory-based ASICs and FPGAs.
-
Efficient regular Expression Evaluation: theory to practice
Proceedings of the 4th ACM/IEEE Symposium on Architectures for Networking and Communications Systems, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been proposed recently to accelerate regular Expression matching and enable deep packet inspection at line rate. This work aims to provide a comprehensive practical Evaluation of existing techniques, extending them and analyzing their compatibility. The study focuses on two hardware architectures: memory-based ASICs and FPGAs.
-
Efficient Regular Expression Evaluation: from Theory to Practice
Symposium on Architectures for Networking and Communications Systems, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been recently proposed in order to\naccelerate regular Expression matching and enable deep packet inspection at\nline rate. This work aims at providing a comprehensive practical Evaluation\nof existing techniques, extending them and analyzing their parallel\napplicability. The study focuses on two different hardware architectures:\nmemory centric systems and FPGA.
Michela Becchi - One of the best experts on this subject based on the ideXlab platform.
-
Data structures, algorithms and architectures for efficient regular Expression Evaluation
2020Co-Authors: Patrick Crowley, Michela BecchiAbstract:Regular Expression matching is a crucial task in several networking applications, where packet payloads need to be inspected at line rate over large sets of complex patterns. Modern, payload-scanning firewalls and intrusion detection systems are perhaps the most notable applications that rely on this mechanism. When targeting memory-based architectures, in which the pattern data structure is stored in off-chip memory such as DRAM, the design of efficient regular Expression matching engines is subject to two basic resource requirements: memory size and memory bandwidth. When designing Field Programmable Gate Array implementations, in which the pattern data structure is stored in on-chip memories or logic gates, the focus is on minimizing logic cell utilization and allowing high operational dock frequency. Keeping these requirements feasible is a challenge as pattern sets grow in size and expressiveness. In this work, we describe novel data structures and algorithms for high-speed regular Expression Evaluation and map the proposed schemes onto suitable parallel architectures. We aim to provide a comprehensive solution that is scalable with respect to both the size of the rule-set and the complexity of the individual rules. Our main contributions are the following. First, we propose an efficient and low complexity compression scheme for deterministic finite automata (DFAs), leading to a compression factor in excess of 98% across a set of representative data sets. Second, we identify and address problematic subpatterns, such as bounded and unbounded repetitions of wildcards and large character sets and back-references. Specifically, we propose a novel automaton with a limited memory footprint and a bounded memory bandwidth requirement that overcomes the limitations of pure DFA-based solutions. Third, we describe how to utilize our proposed techniques when targeting FPGA implementations. Finally, our tools for constructing and analyzing a range of alternative automata for regular-Expression Evaluation have been released as open-source, and are used actively by many researchers from around the world.
-
a dfa a time and space efficient dfa compression algorithm for fast regular Expression Evaluation
ACM Transactions on Architecture and Code Optimization, 2013Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Modern network intrusion detection systems need to perform regular Expression matching at line rate in order to detect the occurrence of critical patterns in packet payloads. While Deterministic Finite Automata (DFAs) allow this operation to be performed in linear time, they may exhibit prohibitive memory requirements. Kumar et al. [2006a] have proposed Delayed Input DFAs (D2FAs), which provide a trade-off between the memory requirements of the compressed DFA and the number of states visited for each character processed, which in turn affects the memory bandwidth required to evaluate regular Expressions. In this article we introduce Amortized time − bandwidth overhead DFAs (A − DFAs), a general compression technique that results in at most N(k + 1)/k state traversals when processing a string of length N, k being a positive integer. In comparison to the D2FA approach, our technique achieves comparable levels of compression with lower provable bounds on memory bandwidth (or greater compression for a given bandwidth bound). Moreover, the A-DFA algorithm has lower complexity, can be applied during DFA creation, and is suitable for scenarios where a compressed DFA needs to be dynamically built or updated. Finally, we show how to combine A-DFA with alphabet reduction and multistride DFAs, two techniques aimed at reducing the memory space and bandwidth requirement of DFAs, and discuss memory encoding schemes suitable for A-DFAs.
-
ANCS - Efficient regular Expression Evaluation: theory to practice
Proceedings of the 4th ACM IEEE Symposium on Architectures for Networking and Communications Systems - ANCS '08, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been proposed recently to accelerate regular Expression matching and enable deep packet inspection at line rate. This work aims to provide a comprehensive practical Evaluation of existing techniques, extending them and analyzing their compatibility. The study focuses on two hardware architectures: memory-based ASICs and FPGAs.
-
Efficient regular Expression Evaluation: theory to practice
Proceedings of the 4th ACM/IEEE Symposium on Architectures for Networking and Communications Systems, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been proposed recently to accelerate regular Expression matching and enable deep packet inspection at line rate. This work aims to provide a comprehensive practical Evaluation of existing techniques, extending them and analyzing their compatibility. The study focuses on two hardware architectures: memory-based ASICs and FPGAs.
-
Efficient Regular Expression Evaluation: from Theory to Practice
Symposium on Architectures for Networking and Communications Systems, 2008Co-Authors: Michela Becchi, Patrick CrowleyAbstract:Several algorithms and techniques have been recently proposed in order to\naccelerate regular Expression matching and enable deep packet inspection at\nline rate. This work aims at providing a comprehensive practical Evaluation\nof existing techniques, extending them and analyzing their parallel\napplicability. The study focuses on two different hardware architectures:\nmemory centric systems and FPGA.
Jinghong Li - One of the best experts on this subject based on the ideXlab platform.
-
sensitive electrochemical aptamer biosensor for dynamic cell surface n glycan Evaluation featuring multivalent recognition and signal amplification on a dendrimer graphene electrode interface
Analytical Chemistry, 2014Co-Authors: Xiaojiao Chen, Yangzhong Wang, Zhuhai Chen, Youyu Zhang, Z Z Li, Jinghong LiAbstract:We demonstrate a multivalent recognition and highly selective aptamer signal amplification strategy for electrochemical cytosensing and dynamic cell surface N-glycan Expression Evaluation by the combination of concanavalin A (Con A), a mannose binding protein, as a model, conjugated poly(amidoamine) dendrimer on a chemically reduced graphene oxide (rGO–DEN) interface, and aptamer- and horseradish peroxidase-modified gold nanoparticles (HRP–aptamer–AuNPs) as nanoprobes. In this strategy, the rGO–DEN can not only enhance the electron transfer ability but also provide a multivalent recognition interface for the conjugation of Con A that avoids the weak carbohydrate–protein interaction and dramatically improves the cell capture efficiency and the sensitivity of the biosensor for cell surface glycan. The high-affinity aptamer- and HRP-modified gold nanoparticles provide an ultrasensitive electrochemical probe with excellent specificity. As proof-of-concept, the detection of CCRF-CEM cell (human acute lymphobla...
-
a functional glycoprotein competitive recognition and signal amplification strategy for carbohydrate protein interaction profiling and cell surface carbohydrate Expression Evaluation
Nanoscale, 2013Co-Authors: Yangzhong Wang, Zhuhai Chen, Jinghong LiAbstract:A simple and sensitive carbohydrate biosensor has been suggested as a potential tool for accurate analysis of cell surface carbohydrate Expression as well as carbohydrate-based therapeutics for a variety of diseases and infections. In this work, a sensitive biosensor for carbohydrate–lectin profiling and in situ cell surface carbohydrate Expression was designed by taking advantage of a functional glycoprotein of glucose oxidase acting as both a multivalent recognition unit and a signal amplification probe. Combining the gold nanoparticle catalyzed luminol electrogenerated chemiluminescence and nanocarrier for active biomolecules, the number of cell surface carbohydrate groups could be conveniently read out. The apparent dissociation constant between GOx@Au probes and Con A was detected to be 1.64 nM and was approximately 5 orders of magnitude smaller than that of mannose and Con A, which would arise from the multivalent effect between the probe and Con A. Both glycoproteins and gold nanoparticles contribute to the high affinity between carbohydrates and lectin. The as-proposed biosensor exhibits excellent analytical performance towards the cytosensing of K562 cells with a detection limit of 18 cells, and the mannose moieties on a single K562 cell were determined to be 1.8 × 1010. The biosensor can also act as a useful tool for antibacterial drug screening and mechanism investigation. This strategy integrates the excellent biocompatibility and multivalent recognition of glycoproteins as well as the significant enzymatic catalysis and gold nanoparticle signal amplification, and avoids the cell pretreatment and labelling process. This would contribute to the glycomic analysis and the understanding of complex native glycan-related biological processes.
F. Coelho - One of the best experts on this subject based on the ideXlab platform.
-
IEEE PACT - Using algebraic transformations to optimize Expression Evaluation in scientific code
Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192), 1998Co-Authors: J. Zory, F. CoelhoAbstract:Algebraic properties such as associativity or distributivity allow the manipulation of a set of mathematically equivalent Expressions. However as shown in this paper the cost of evaluating such Expressions on a computer is not constant within this domain. We suggest the use of algebraic transformations to improve the performance of computationally intensive applications on modern computer architectures. We claim that taking into account instruction-level parallelism and the new capabilities of processors when applying these transformations leads to large run-time improvements. Due to a combinatorial explosion, associative commutative pattern-matching techniques cannot systematically be used in this context. Thus, we introduce two performance enhancing algorithms providing factorization and multiply-add extraction heuristics and choice criteria based on a simple cost model. This paper describes our approach and a first implementation. Experiments on real code, including an excerpt from SPEC FP95, are very promising since we automatically obtain the same results as manual transformations, with a performance improvement by a factor of up to three.
-
Using algebraic transformations to optimize Expression Evaluation in scientific code
Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192), 1998Co-Authors: J. Zory, F. CoelhoAbstract:Algebraic properties such as associativity or distributivity allow the manipulation of a set of mathematically equivalent Expressions. However as shown in this paper the cost of evaluating such Expressions on a computer is not constant within this domain. We suggest the use of algebraic transformations to improve the performance of computationally intensive applications on modern computer architectures. We claim that taking into account instruction-level parallelism and the new capabilities of processors when applying these transformations leads to large run-time improvements. Due to a combinatorial explosion, associative commutative pattern-matching techniques cannot systematically be used in this context. Thus, we introduce two performance enhancing algorithms providing factorization and multiply-add extraction heuristics and choice criteria based on a simple cost model. This paper describes our approach and a first implementation. Experiments on real code, including an excerpt from SPEC FP95, are very promising since we automatically obtain the same results as manual transformations, with a performance improvement by a factor of up to three.
Amruth N. Kumar - One of the best experts on this subject based on the ideXlab platform.
-
ITiCSE - The Effectiveness of Visualization for Learning Expression Evaluation: A Reproducibility Study
Proceedings of the 2016 ACM Conference on Innovation and Technology in Computer Science Education - ITiCSE '16, 2016Co-Authors: Amruth N. KumarAbstract:A study was conducted to reproduce the results of an earlier study on the effectiveness of visualization for learning Expression Evaluation in a problem-solving software tutor on arithmetic Expressions. In the current reproducibility study, data was collected from a software tutor on assignment Expressions over six semesters. ANOVA analysis of the amount and speed of learning was conducted with treatment, sex and racial groups as fixed factors. Results include that visualization helped the students learn significantly more concepts, whether the students needed to use the tutor or benefited from using the tutor. However, it only benefited the less-prepared students. It did not help the students learn faster. It benefited both the sexes and traditionally represented as well as underrepresented groups. The current study confirmed almost all the results from the previous study, albeit for a harder topic. One reason why visualization was found to be effective in both these studies may be that the same visualization scheme was used by the students to both view feedback and construct their answers.
-
SIGCSE - The Effectiveness of Visualization for Learning Expression Evaluation
Proceedings of the 46th ACM Technical Symposium on Computer Science Education - SIGCSE '15, 2015Co-Authors: Amruth N. KumarAbstract:A controlled study was conducted to evaluate the effectiveness of providing visualization as part of feedback in a problem-solving software tutor on arithmetic Expression Evaluation. Data was collected over six semesters from multiple institutions. ANOVA analysis of the collected data was conducted in three stages. Statistically significant results include that visualization helped students learn more concepts; visualization did not improve the speed of learning; the benefits of visualization accrued primarily to less-prepared students; and visualization may affect different demographic subgroups differently. Incidental results include that there was no difference among demographic groups (male/female, traditional/underrepresented, Computer Science/non-CS) in the number of concepts learned using the software tutor, although some groups (female, underrepresented) were less-prepared before using the tutor than their counterparts, and some groups learned concepts with fewer practice problems (male, traditionally represented) than their counterparts. Concurrence of the results obtained whether the analysis was conducted based on pre-condition (need) or post-condition (benefit) of using the tutor strengthens the claims made as a result of this study.
-
AIED - The Effects of Error-Flagging in a Tutor on Expression Evaluation
2007Co-Authors: Amruth N. Kumar, Peter RutiglianoAbstract:We evaluated the effect of providing error-flagging as support for error detection, but not error correction while the student is solving a problem. We found that providing error-flagging in addition to demand feedback during practice learning was no more effective than providing only demand feedback when the tutor did not explicitly mention that errors were being flagged. On the other hand, explaining and providing error-flagging without demand feedback during pre-and post-tests resulted in significantly better scores on pre-and post-tests even though error-flagging did not provide any error-correction support.
-
SIGCSE - Results from the Evaluation of the effectiveness of an online tutor on Expression Evaluation
ACM SIGCSE Bulletin, 2005Co-Authors: Amruth N. KumarAbstract:Researchers have been developing online tutors for various disciplines, including Computer Science. Educators are increasingly using online tutors to supplement their courses. Are online tutors effective? Can they help students learn? If so, what features contribute to their effectiveness? We will examine these questions in the context of an online tutor that we developed for introductory Computer Science. The tutor is designed to help students learn Expression Evaluation in C++/Java.We evaluated the tutor over several years, in multiple sections of Computer Science I each year. We used controlled tests with differential treatments, and used pre and post-tests to evaluate the effectiveness of the tutor. Our results show that online tutors indeed help students learn. Students who use the tutor for practice learn better than those who use a printed workbook. Students who receive both graphic visualization and text explanation learn better than those who receive only graphic visualization. Students who use graphic visualization learn better than those who receive no explanation. These results will be of interest to both developers and users of online tutors.
-
A problem generator to learn Expression: Evaluation in CSI, and its effectiveness
Journal of Computing Sciences in Colleges, 2001Co-Authors: Aravind K. Krishna, Amruth N. KumarAbstract:Introductory Computer Science students often do not clearly understand the concepts of operator precedence and associativity, and make mistakes when they write and evaluate Expressions. In order to help them learn these concepts by solving problems, we have developed an applet that generates problems (henceforth called problet) and corrects the learner's answers. In this paper, we will describe the design and implementation of this problet. We conducted a controlled experiment to assess the effectiveness of using the problet. Results indicate that using the problet improved learning by 18%. We will describe our experiment and present its results.
