WARD: a transmission control protocol-friendly stateless active queue management scheme

In this article, the problem of providing a fair bandwidth allocation to the flows sharing a congested link in a router is investigated. Queue management, bandwidth share and congestion control are very important to both the robustness and fairness of the Internet. The buffer at the outgoing link is a simple FIFO, shared by packets belonging to the flows. A new transmission control protocol (TCP)-friendly router-based active queue management scheme, termed WARD, is proposed to approximate the fair queueing policy. WARD is a simple packet-dropping algorithm with a random mechanism which discriminates against flows that submit more packets per second than is allowed as their fair share. By doing this, it not only protects TCP connections from user datagram protocol flows, but also solves the problem of competing bandwidth among different TCP versions, such as TCP Vegas and TCP Reno. In addition, WARD works quite well for TCP flow isolation even with different round trip times. In other words, WARD improves the unfair bandwidth allocation properties. Furthermore, as it is stateless and easy to implement, WARD controls unresponsive or misbehaving flows with only a minimum overhead.     

水乡城市滨水景观设计初探——以浙江省嘉兴为例

滨水是指与河流、湖泊、江海毗邻的土地或建筑物,它既是陆的边沿,也是水的边缘。作为一种特殊的城市用地,城市滨水区是城市中具有巨大发展潜力的空间领域。     

An efficient resizing technique for the design of tall steel buildings subject to multiple drift constraints

This paper presents an efficient resizing technique for the optimum design of tall steel building frameworks. Specifically, an ‘optimality criteria’ method is applied to minimize the weight of a lateral load-resisting structural system of fixed topology subject to constraints on interstorey drift. By exploiting the fact for building frameworks that member forces are relatively insensitive to changes in member sizes, rigorously-derived optimality criteria are shown to be readily satisfied through an iterative pseudo-discrete optimization procedure that converges in but a few cycles to a least-weight design using commercial-standard steel sections. While not considered herein, it is a simple matter to extend the stiffness-based design procedure to account also for strength requirements. Two building framework examples are presented to illustrate the features of the design optimization method.     

Cross sections of proton induced nuclear reactions on enriched 111Cd and 112Cd for the production of 111In for use in nuclear medicine

Proton induced nuclear reactions on enriched 111Cd and 112Cd have been studied up to 30 MeV in the context of routine production of the medically used isotope 111In with low and medium energy cyclotrons. The excitation functions of 111Cd(p,n)111m,gIn and 112Cd(p,2n)111m,gIn as production reactions and 111Cd(p,2n)110mIn, 111Cd(p,2n)110In, 111Cd(p,3n)109ml,m2,In, 112Cd(p,3n)110mIn, 112Cd(p,3n)110gIn as competing processes have been measured using the activation method involving the stacked-foil technique. The deduced thick target yields are compared with those obtained experimentally.     

An overview of electric vehicle technology

An overview of the present status and future trends in electric vehicle technology is provided. The emphasis is on the impact of rapid development of electric motors, power electronics, microelectronics, and new materials. Comparisons are made among various electric drive systems and battery systems. The market size of electric vehicles in the coming years and the potential electric vehicle impacts are discussed     

A knowledge-based expert system for gearing design application using Prolog and C

This paper describes the development of an expert system in gearing design application. The knowledge in this design field is vast and involves selection of appropriate data from large amounts of engineering standards data available in engineering catalogues and design handbooks. Experts very oftens have different opinions in determining the design solution. The theme of the work is to organise, systemise, conserve and spread this knowledge relating to gearing design and configuration.

There is a large number of design parameters involved in gearing design. The design of gears also requires an iterative approach to optimize these design parameters that govern the kinematic, as well as the strength performance. Owing to the complex combinations of these aspects, conventional design office practice tends to become complicated and time consuming. Hence, in this piece of work, attempts are made to formulate a design system capable of designing a required type of gear and its configuration under a specified set of working conditions.

The main features of the expert gearing design system are the use of semantic nets in representing the objects, a frame-based data acquisition and representation system, and production rules in representing the domain-specific knowledge base. The system provides a very user-friendly interface through the use of an expert system development tool Prolog. While Prolog is used to set up the various databases and the inference engine, the detail gear design calculations are performed by C. It is found that the communication between these two language media is excellent in terms of performance and efficiency.     

Role of imidazoline receptors in the cardiovascular actions of moxonidine, rilmenidine and clonidine in conscious rabbits

The present study in conscious rabbits with intracisternal (i.c.) catheters sought to determine the relative contribution of the I1 subtype of imidazoline receptors (IR) and alpha 2 adrenoceptors to the hypotensive effects of rilmenidine, clonidine and moxonidine with an I1-IR/alpha 2 adrenoceptor antagonist efaroxan and a specific alpha 2 adrenoceptor antagonist 2-methoxyidazoxan (2-MI). The alpha 2 adrenoceptor antagonist effect of efaroxan was compared with 2-MI by performing cumulative dose-response curves in the presence of alpha-methyldopa (400 micrograms/kg i.c.). 2-MI was 5.6 times more potent than efaroxan at reversing 75% of the hypotension elicited by alpha-methyldopa (P < .025). This dose ratio was used to match doses of efaroxan and 2-MI for similar alpha 2 adrenoceptor blockade. The effects of efaroxan (4.1, 13, 41 micrograms/kg i.c.) and 2-MI (0.74, 2.3, 7.4 micrograms/kg i.c.) were investigated on a single i.c. dose of rilmenidine (12 micrograms/kg), clonidine (0.75 microgram/kg) and moxonidine (0.51 microgram/kg). These doses of the antihypertensive agents, which were determined from cumulative dose-response curves, produce 90% of the maximum hypotension. Efaroxan was more effective at reversing the hypotension induced by moxonidine and rilmenidine than was 2-MI (P < .01). These findings suggest that rilmenidine and moxonidine act predominantly via IR. By contrast, 2-MI was more effective at reversing the clonidine-induced hypotension than was efaroxan (P < .001), suggesting that clonidine acts mainly via alpha 2 adrenoceptors in conscious normotensive rabbits. Thus, a higher selectivity of the second generation agents moxonidine and rilmenidine for I1-IR over alpha 2 adrenoceptors, compared with the first generation agent clonidine, appears to be necessary for this effect to be manifested in their hypotensive actions.     

Clinical presentation and prognostic factors in sodium monofluoroacetate intoxication

From 1970 to 1992 a total of 63 patients underwent operation for ampullary tumor: 40 pancreatoduodenectomies (PDs), 3 total PDs, 8 ampullectomies, and 12 bypass or exploratory laparotomies. The resectability rate was 68%. There were 9 benign tumors, 1 anaplastic tumor, and 53 adenocarcinomas. According to Martin's classification, there were 7 stage I, 11 stage II, 14 stage III, and 21 stage IV tumors. All patients with stage I, II, and III tumors underwent resection. Patients with stage IV tumors had either resection (n = 11) or bypass (n = 10). The mean duration of hospital stay was 20.6 days. Operative mortality was 12.7% for the whole series and 7.5% after PD (2.5% for the last 10 years). Overall survival was 40% at 5 years (85% for stage I, 65% for stage II, 44% for stage III, and 8% for stage IV). Survival was better for stages I, II, and III after PD than after ampullectomy. For stage IV patients survival was 70% after PD versus 20% after bypass at 1 year and 25% versus 0% after 2 years. In our opinion, PD should be proposed even for benign lesions because two of our patients had to undergo repeat operation (PD) 4 and 22 years later, respectively, for stage IV disease. PD is our choice for all tumors of the ampulla.