Packet classification by multi-level cutting of the classification space: An algorithmic-architectural solution for IP packet classification in next generation networks

Traditionally, the Internet provides only a “best-effort” service, treating all packets going to the same destination equally. However, providing differentiated services for different users based on their quality requirements is increasingly becoming a demanding issue. For this, routers need to have the capability to distinguish and isolate traffic belonging to different flows. This ability to determine the flow each packet belongs to is called packet classification. Technology vendors are reluctant to support algorithmic solutions for classification due to their non-deterministic performance. Although CAMs are favoured by technology vendors due to their deterministic high lookup rates, they suffer from the problems of high power dissipation and high silicon cost. This paper provides a new algorithmic-architectural solution for packet classification that mixes CAMs with algorithms based on multi-level cutting the classification space into smaller spaces. The provided solution utilizes the geometrical distribution of rules in the classification space. It provides the deterministic performance of CAMs, support for dynamic updates, and added flexibility for system designers.

A Novel Packet Marking Function for Real-Time Interactive MPEG-4 Video Applications in a Differentiated Services Network

The future convergence of voice, video and data applications on the Internet requires that next generation technology provides bandwidth and delay guarantees. Current technology trends are moving towards scalable aggregate-based systems where applications are grouped together and guarantees are provided at the aggregate level only. This solution alone is not enough for interactive video applications with sub-second delay bounds. This paper introduces a novel packet marking scheme that controls the end-to-end delay of an individual flow as it traverses a network enabled to supply aggregate- granularity Quality of Service (QoS). IPv6 Hop-by-Hop extension header fields are used to track the packet delay encountered at each network node and autonomous decisions are made on the best queuing strategy to employ. The results of network simulations are presented and it is shown that when the proposed mechanism is employed the requested delay bound is met with a 20% reduction in resource reservation and no packet loss in the network.

Fairness measurement for adaptive packet schedulers

An equation is presented for calculating the fairness of dynamically adaptive packet schedulers such as dynamic weighted fair queuing (DWFQ). The fairness of static packet schedulers such as weighted fair queue (WFQ) can be found using the widely accepted Worst-case Fair Index. The fairness of DWFQ can be measured using an Adapted Worst-case Fairness Index (AWFI). The AWFI enables a direct comparison of fairness properties of the DWFQ or other dynamically adaptive schedulers with static/non-adaptive schedulers.

Steam reforming of model compounds and fast pyrolysis bio-oil on supported noble metal catalysts

The production of hydrogen by steam reforming of bio-oils obtained from the fast pyrolysis of biomass requires the development of efficient catalysts able to cope with the complex chemical nature of the reactant. The present work focuses on the use of noble metal-based catalysts for the steam reforming of a few model compounds and that of an actual bio-oil. The steam reforming of the model compounds was investigated in the temperature range 650-950 degrees C over Pt, Pd and Rh supported on alumina and a ceria-zirconia sample. The model compounds used were acetic acid, phenol, acetone and ethanol. The nature of the support appeared to play a significant role in the activity of these catalysts. The use of ceria-zirconia, a redox mixed oxide, lead to higher H-2 yields as compared to the case of the alumina-supported catalysts. The supported Rh and Pt catalysts were the most active for the steam reforming of these compounds, while Pd-based catalysts poorly performed. The activity of the promising Pt and Rh catalysts was also investigated for the steam reforming of a bio-oil obtained from beech wood fast pyrolysis. Temperatures close to, or higher than, 800 degrees C were required to achieve significant conversions to COx and H-2 (e.g., H-2 yields around 70%). The ceria-zirconia materials showed a higher activity than the corresponding alumina samples. A Pt/ceria-zirconia sample used for over 9 h showed essentially constant activity, while extensive carbonaceous deposits were observed on the quartz reactor walls from early time on stream. In the present case, no benefit was observed by adding a small amount of O-2 to the steam/bio-oil feed (autothermal reforming, ATR), probably partly due to the already high concentration of oxygen in the bio-oil composition. (c) 2005 Elsevier B.V. All rights reserved.

The preparation and properties of coprecipitated Cu-Zr-Y and Cu-Zr-La catalysts used for the steam reforming of methanol

A series of Cu-zirconia catalysts containing various additives (Y2O3, La2O3, Al2O3 and CeO2) have been prepared by coprecipitation and their activities and stabilities under operating conditions have been obtained for the steam reforming of methanol. It has been found that an yttria-promoted catalyst containing 30 mol% Cu and 20 mol% of Y2O3 is not only very active but is also very stable under reaction conditions. The yttria appears to stabilise a high copper surface area and may also have a slight promotional effect on the copper. The results obtained with this material compare very favourably with data for the best catalysts reported in the literature. (C) 2007 Published by Elsevier B.V.

Packet classification by multi-level cutting of the classification space

Traditionally, the Internet provides only a “best-effort” service, treating all packets going to the same destination equally. However, providing differentiated services for different users based on their quality requirements is increasingly becoming a demanding issue. For this, routers need to have the capability to distinguish and isolate traffic belonging to different flows. This ability to determine the flow each packet belongs to is called packet classification. Technology vendors are reluctant to support algorithmic solutions for classification due to their nondeterministic performance. Although content addressable memories (CAMs) are favoured by technology vendors due to their deterministic high-lookup rates, they suffer from the problems of high-power consumption and high-silicon cost. This paper provides a new algorithmic-architectural solution for packet classification that mixes CAMs with algorithms based on multilevel cutting of the classification space into smaller spaces. The provided solution utilizes the geometrical distribution of rules in the classification space. It provides the deterministic performance of CAMs, support for dynamic updates, and added flexibility for system designers.