ARM-based SoC design and emulation of a compressed bit-vector packet classification algorithm
Bibliographic record
Abstract
Packet classification (PC) is the problem of matching incoming packets to a router against a database of rules or filters.The rules speciff a directive for incoming packets and provide a means of implementing new services such as Quality of Service (QoS) guarantees.This thesis presents state-of-the-art developments and emerging approaches to packet classification along with a set of requirements to meet the demands of existing and future large bandwidth connections.Although many schemes have been proposed to solve the multldimensional pC problem, none of them scale well beyond two dimensions in terms of speed and,/or rule size.Based on this evaluation, a novel algorithm is proposed.The new and ìnnovative algorithm builds upon two geometric hardware-based approaches, namely the Lucent Bit-I/ector (BY) and, Aggegated Bit-I/ector (ABV) algorithms t1l, t2l.Two new ideas are presented, creating what shall be called lhe Compressed Bit-Vector (CBY) algorithm.This CBV scheme employs a divide-and-conquer technique breaking the general multi-dimensional problem into more manageable two-dimensional problems each producing compressed bit-vector solutions.This provides a framework that will allow for the incorporation of various two-dimensional search algorithms into one system that can perform multi- dimensional classification.Similar to the ABV and Lucent BV schemes, the CBV algorithm is amenable to a hardware and software implementation.As a proof-of-concept, a portion ofthe algorithm is implemented and functionally verified on a rapid prototyping platþrm (RPP) by adherin gto a System-on-Chþ (SoC) design flow.Through this process, it is further demonstrated how platform-based design provides many benefits without a cornmitment to silicon.It allows for embedded software development early in the design cycle, the ¡esolution of hardware and software integration problems, the ability to validate third party IP blocks in a system environment, and the capability to experim€nt with the inplementation.It is also useful for acquiring realistic performance statistics to determi¡e if application-speci,fic integrated circuit (ASIC) performance targets are achievable.Using the platform, an extensive performance analysis ofthe CBV algorithm is achieved on synthetic rule- sets modeled after real firewall database statistics.Out of this study, it is shown that the algorithm's memory usage is scalable to large rule-sets.Recommendations are also made on how an ASIC implementation could sustana Gigabit Ethernet (GE) li¡e-rate on average size packets.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".