Packet Reordering in Network Processors

Network processors today consist of multiple parallel processors (micro engines) with support for multiple threads to exploit packet level parallelism inherent in network workloads. With such concurrency, packet ordering at the output of the network processor cannot be guaranteed. This paper studies the effect of concurrency in network processors on packet ordering. We use a validated Petri net model of a commercial network processor, Intel IXP 2400, to determine the extent of packet reordering for IPv4 forwarding application. Our study indicates that in addition to the parallel processing in the network processor, the allocation scheme for the transmit buffer also adversely impacts packet ordering. In particular, our results reveal that these packet reordering results in a packet retransmission rate of up to 61%. We explore different transmit buffer allocation schemes namely, contiguous, strided, local, and global which reduces the packet retransmission to 24%. We propose an alternative scheme, packet sort, which guarantees complete packet ordering while achieving a throughput of 2.5 Gbps. Further, packet sort outperforms the in-built packet ordering schemes in the IXP processor by up to 35%.

Exploiting Programmable Network Interfaces for Parallel Query Execution in Workstation Clusters

Workstation clusters equipped with high performance interconnect having programmable network processors facilitate interesting opportunities to enhance the performance of parallel application run on them. In this paper, we propose schemes where certain application level processing in parallel database query execution is performed on the network processor. We evaluate the performance of TPC-H queries executing on a high end cluster where all tuple processing is done on the host processor, using a timed Petri net model, and find that tuple processing costs on the host processor dominate the execution time. These results are validated using a small cluster. We therefore propose 4 schemes where certain tuple processing activity is offloaded to the network processor. The first 2 schemes offload the tuple splitting activity - computation to identify the node on which to process the tuples, resulting in an execution time speedup of 1.09 relative to the base scheme, but with I/O bus becoming the bottleneck resource. In the 3rd scheme in addition to offloading tuple processing activity, the disk and network interface are combined to avoid the I/O bus bottleneck, which results in speedups up to 1.16, but with high host processor utilization. Our 4th scheme where the network processor also performs apart of join operation along with the host processor, gives a speedup of 1.47 along with balanced system resource utilizations. Further we observe that the proposed schemes perform equally well even in a scaled architecture i.e., when the number of processors is increased from 2 to 64

Performance modeling and architecture exploration of network processors

This paper proposes a Petri net model for a commercial network processor (Intel iXP architecture) which is a multithreaded multiprocessor architecture. We consider and model three different applications viz., IPv4 forwarding, network address translation, and IP security running on IXP 2400/2850. A salient feature of the Petri net model is its ability to model the application, architecture and their interaction in great detail. The model is validated using the Intel proprietary tool (SDK 3.51 for IXP architecture) over a range of configurations. We conduct a detailed performance evaluation, identify the bottleneck resource, and propose a few architectural extensions and evaluate them in detail.

Evaluation of dynamic voltage and frequency scaling for stream programs

Dynamic Voltage and Frequency Scaling (DVFS) offers a huge potential for designing trade-offs involving energy, power, temperature and performance of computing systems. In this paper, we evaluate three different DVFS schemes - our enhancement of a Petri net performance model based DVFS method for sequential programs to stream programs, a simple profile based Linear Scaling method, and an existing hardware based DVFS method for multithreaded applications - using multithreaded stream applications, in a full system Chip Multiprocessor (CMP) simulator. From our evaluation, we find that the software based methods achieve significant Energy/Throughput2(ET−2) improvements. The hardware based scheme degrades performance heavily and suffers ET−2 loss. Our results indicate that the simple profile based scheme achieves the benefits of the complex Petri net based scheme for stream programs, and present a strong case for the need for independent voltage/frequency control for different cores of CMPs, which is lacking in most of the state-of-the-art CMPs. This is in contrast to the conclusions of a recent evaluation of per-core DVFS schemes for multithreaded applications for CMPs.

Performance modelling of a fault-tolerant real-time multiprocessor using stochastic Petri nets

The fault-tolerant multiprocessor (ftmp) is a bus-based multiprocessor architecture with real-time and fault- tolerance features and is used in critical aerospace applications. A preliminary performance evaluation is of crucial importance in the design of such systems. In this paper, we review stochastic Petri nets (spn) and developspn-based performance models forftmp. These performance models enable efficient computation of important performance measures such as processing power, bus contention, bus utilization, and waiting times.

On the invariants of coloured Petri Nets

In this paper, we develop a theorem that enables computation of the place invariants of the union of a finite collection of coloured Petri Nets when the individual nets satisfy certain conditions and their invariants are known. We consider the illustrative examples of the Readers-Writers problem, a resource sharing system, and a network of databases and show how this theorem is a valuable tool in the analysis of concurrent systems.

A hydro-wood net-energy approach to hydro project design

An integrated approach to energy planning, when applied to large hydroelectric projects, requires that the energy-opportunity cost of the land submerged under the reservoir be incorporated into the planning methodology. Biomass energy lost from the submerged land has to be compared to the electrical energy generated, for which we develop four alternative formulations of the net-energy function. The design problem is posed as an LP problem and is solved for two sites in India. Our results show that the proposed designs may not be viable in net-energy terms, whereas a marginal reduction in the generation capacity could lead to an optimal design that gives substantial savings in the submerged area. Allowing seasonal variations in the hydroelectric generation capacity also reduces the reservoir size. A mixed hydro-wood generation system is then examined and is found to be viable.

COMMON-Sense net:Improved water management for resource-poor farmers via sensor networks

We describe the on-going design and implementation of a sensor network for agricultural management targeted at resource-poor farmers in India. Our focus on semi-arid regions led us to concentrate on water-related issues. Throughout 2004, we carried out a survey on the information needs of the population living in a cluster of villages in our study area. The results highlighted the potential that environment-related information has for the improvement of farming strategies in the face of highly variable conditions, in particular for risk management strategies (choice of crop varieties, sowing and harvest periods, prevention of pests and diseases, efficient use of irrigation water etc.). This leads us to advocate an original use of Information and Communication Technologies (ICT). We believe our demand-driven approach for the design of appropriate ICT tools that are targeted at the resource-poor to be relatively new. In order to go beyond a pure technocratic approach, we adopted an iterative, participatory methodology.

Stochastic models of IEEE 802.11e wireless net works with multimedia applications

We provide a survey of some of our recent results ([9], [13], [4], [6], [7]) on the analytical performance modeling of IEEE 802.11 wireless local area networks (WLANs). We first present extensions of the decoupling approach of Bianchi ([1]) to the saturation analysis of IEEE 802.11e networks with multiple traffic classes. We have found that even when analysing WLANs with unsaturated nodes the following state dependent service model works well: when a certain set of nodes is nonempty, their channel attempt behaviour is obtained from the corresponding fixed point analysis of the saturated system. We will present our experiences in using this approximation to model multimedia traffic over an IEEE 802.11e network using the enhanced DCF channel access (EDCA) mechanism. We have found that we can model TCP controlled file transfers, VoIP packet telephony, and streaming video in the IEEE802.11e setting by this simple approximation.

A new methodology for analyzing distributed systems modeled by petri nets

Distributed computing systems can be modeled adequately by Petri nets. The computation of invariants of Petri nets becomes necessary for proving the properties of modeled systems. This paper presents a two-phase, bottom-up approach for invariant computation and analysis of Petri nets. In the first phase, a newly defined subnet, called the RP-subnet, with an invariant is chosen. In the second phase, the selected RP-subnet is analyzed. Our methodology is illustrated with two examples viz., the dining philosophers' problem and the connection-disconnection phase of a transport protocol. We believe that this new method, which is computationally no worse than the existing techniques, would simplify the analysis of many practical distributed systems.

Multi-decadal variation of the net downward shortwave radiation over south Asia: The solar dimming effect

The solar radiation flux at the earth's surface has gone through decadal changes of decreasing and increasing trends over the globe. These phenomena known as dimming and brightening, respectively, have attracted the scientific interest in relation to the changes in radiative balance and climate. Despite the interest in the solar dimming/brightening phenomenon in various parts of the world, south Asia has not attracted great scientific attention so far. The present work uses the net downward shortwave radiation (NDSWR) values derived from satellites (Modern Era Retrospective-analysis for Research and Applications, MERRA 2D) in order to examine the multi-decadal variations in the incoming solar radiation over south Asia for the period of 1979-2004. From the analysis it is seen that solar dimming continues over south Asia with a trend of -0.54 Wm(-2) yr(-1). Assuming clear skies an average decrease of -0.05 Wm(-2)yr(-1) in NDSWR was observed, which is attributed to increased aerosol emissions over the region. There is evidence that the increase in cloud optical depth plays the major role for the solar dimming over the area. The cloud optical depth (MERRA retrievals) has increased by 10.7% during the study period, with the largest increase to be detected for the high-level (atmospheric pressure P < 400 hPa) clouds (31.2%). Nevertheless, the decrease in solar radiation and the role of aerosols and clouds exhibit large monthly and seasonal variations directly affected by the local monsoon system, the anthropogenic and natural aerosol emissions. All these aspects are examined in detail aiming at shedding light into the solar dimming phenomenon over a densely populated area. (C) 2011 Elsevier Ltd. All rights reserved.

Effect of Net-Fluence on waveguide formation in ultrafast laser inscribed chalcogenide glass

Waveguides were fabricated on GeGaSEr chalcogenide glass using ultrafast laser inscription method. The thermal diffusion model is discussed for understanding the light matter interaction and shown the effect of net-fluence in waveguide formation on chalcogenide glass. (C) 2012 Optical Society of America

Large-Scale Elastic Net Regularized Linear Classification SVMs and Logistic Regression

Elastic Net Regularizers have shown much promise in designing sparse classifiers for linear classification. In this work, we propose an alternating optimization approach to solve the dual problems of elastic net regularized linear classification Support Vector Machines (SVMs) and logistic regression (LR). One of the sub-problems turns out to be a simple projection. The other sub-problem can be solved using dual coordinate descent methods developed for non-sparse L2-regularized linear SVMs and LR, without altering their iteration complexity and convergence properties. Experiments on very large datasets indicate that the proposed dual coordinate descent - projection (DCD-P) methods are fast and achieve comparable generalization performance after the first pass through the data, with extremely sparse models.

Stabilization of Co-3 - Oxoclusters in a pcu Net: Synthesis, Structure, Solvent Exchange (Single Crystal to Single Crystal) and Magnetic Studies

The solvothermal reaction of CoCl(2)4H(2)O and 4,4-sulfonyldibenzoic acid (H(2)SDBA) resulted in the formation of a three-dimensional coordination polymer Co-3(C14H8O6S)(3)(DMA)(2)(MeOH)].DMA (Ia) consisting of trinuclear Co-3 oxo-cluster units. The Co-3 trimeric units are connected by SDBA(2-) anions leading to a three dimensional structure with a pcu topology. The terminal methanol molecules could be exchanged in a single crystal to single crystal (SCSC) fashion by other similar solvent molecules (ethanol, acetonitrile, water, ethyleneglycol). Magnetic studies on the parent compound, Ia, indicate antiferromagnetic interactions between the central metal atoms.