A Sorted-Partitioning Approach to Fast and Scalable Dynamic Packet Classification

Yingchareonthawornchai, Sorrachai; Daly, James E.; Liu, Alex X.; Torng, Eric

doi:10.1109/tnet.2018.2852710

Cited by 48 publications

(23 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A longer processing latency could be expected. PartitionSort [19] reduces average lookup times by presenting a pre-computed priority for each tuple space, but there is no guarantee of its worst-case performance. The further TupleMerge [20] relaxes the rule's restrictions on entering the same tuple to reduce the number of tuples.…”

Section: Packet Classification Algorithmsmentioning

confidence: 99%

“…Due to its high performance with flexible reconfigurability, FPGA is increasingly being utilized for accelerating SDN software switches [11,12]. Decision-tree-based algorithms [13][14][15][16] and Tuple-Space-Search (TSS) algorithms [17][18][19][20] are two major approaches implemented on the CPU/GPU platform. However, the characteristics of these algorithms conflict with FPGA devices and cannot be effectively migrated to FPGA.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Enabling Packet Classification with Low Update Latency for SDN Switch on FPGA

et al. 2020

Sustainability

View full text Add to dashboard Cite

Field Programmable Gate Array (FPGA) is widely used in real-time network processing such as Software-Defined Networking (SDN) switch due to high performance and programmability. Bit-Vector (BV)-based approaches can implement high-performance multi-field packet classification, on FPGA, which is the core function of the SDN switch. However, the SDN switch requires not only high performance but also low update latency to avoid controller failure. Unfortunately, the update latency of BV-based approaches is inversely proportional to the number of rules, which means can hardly support the SDN switch effectively. It is reasonable to split the ruleset into sub-rulesets that can be performed in parallel, thereby reducing update latency. We thus present SplitBV for the efficient update by using several distinguishable exact-bits to split the ruleset. SplitBV consists of a constrained recursive algorithm for selecting the bit positions that can minimize the latency and a hybrid lookup pipeline. It can achieve a significant reduction in update latency with negligible memory growth and comparable high performance. We implement SplitBV and evaluate its performance by simulation and FPGA prototype. Experimental results show that our approach can reduce 73% and 36% update latency on average for synthetic 5-tuple rules and OpenFlow rules respectively.

show abstract

Section: Packet Classification Algorithmsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Enabling Packet Classification with Low Update Latency for SDN Switch on FPGA

et al. 2020

Sustainability

View full text Add to dashboard Cite

show abstract

“…There are also many schemes that employ decision trees for packet classification. For example, Par-titionSort [21], CutSplit [22], TabTree [23], NeuroCuts [24], CutTSS [25], and NuevoMatch [26] all proposed a certain kind of method to construct a decision tree for fast packet classification. These schemes mainly aim at alleviating the search complexity of the packet classification in with the rules stored in RAM.…”

Section: Related Workmentioning

confidence: 99%

RETCAM: An efficient TCAM compression model for flow table of OpenFlow

Zhang

Sun

et al. 2020

J. Commun. Netw.

View full text Add to dashboard Cite

OpenFlow is a widely adopted dataplane protocol in software-defined networking (SDN). However, the expansion of supported match fields in OpenFlow brings additional pressure to the storage space of ternary content addressable memory (TCAM) in physical device, since the arbitrary wildcard support in the match field of OpenFlow relies heavily on TCAM for looking-up speed. In this paper, a mathematical model aiming at the storage space reduction of the flow table in TCAM is presented, which is named as RETCAM. RETCAM analyzes the relationships among all the match fields and then categorize the redundancy among different fields into three types. Based on the three redundancy types, three compression algorithms named as inter-field merge, field mapping and intra-field compression are presented. The outcomes of each compression algorithm are flow entries with smaller bit-width which is sent to TCAM for flow matching. In this way, the flexibility of OpenFlow is not harmed, thus maintaining the function integrity of the original flow table. Simulation at the end shows that RETCAM saves almost about 60% of TCAM space for a given flow table with no damage to the function integrity of OpenFlow, and the compression performance stands stable with the increase of flow table size.

show abstract

“…However, they cannot support high flexibility to customers’ various demands [7,8,9]. For software-based algorithms, some algorithms build a small search table to accommodate a large rule-set with limited memory [10,11]. Others focus on speeding up the classification process by spending more resources, in order to support vast traffic [12].…”

Section: Introductionmentioning

confidence: 99%

Wildcard Fields-Based Partitioning for Fast and Scalable Packet Classification in Vehicle-to-Everything

Wee

Choi

Pak

2019

Sensors

View full text Add to dashboard Cite

Vehicle-to-Everything (V2X) requires high-speed communication and high-level security. However, as the number of connected devices increases exponentially, communication networks are suffering from huge traffic and various security issues. It is well known that performance and security of network equipment significantly depends on the packet classification algorithm because it is one of the most fundamental packet processing functions. Thus, the algorithm should run fast even with the huge set of packet processing rules. Unfortunately, previous packet classification algorithms have focused on the processing speed only, failing to be scalable with the rule-set size. In this paper, we propose a new packet classification approach balancing classification speed and scalability. It can be applied to most decision tree-based packet classification algorithms such as HyperCuts and EffiCuts. It determines partitioning fields considering the rule duplication explicitly, which makes the algorithm memory-effective. In addition, the proposed approach reduces the decision tree size substantially with the minimal sacrifice of classification performance. As a result, we can attain high-speed packet classification and scalability simultaneously, which is very essential for latest services such as V2X and Internet-of-Things (IoT).

show abstract

A Sorted-Partitioning Approach to Fast and Scalable Dynamic Packet Classification

Cited by 48 publications

References 29 publications

Enabling Packet Classification with Low Update Latency for SDN Switch on FPGA

Enabling Packet Classification with Low Update Latency for SDN Switch on FPGA

RETCAM: An efficient TCAM compression model for flow table of OpenFlow

Wildcard Fields-Based Partitioning for Fast and Scalable Packet Classification in Vehicle-to-Everything

Contact Info

Product

Resources

About