Llorenç Cerdà

Design and implementation of high-performance memory systems for future packet buffers

In this paper, we address the design of a future high-speed router that supports line rates as high as OC-3072 (160 Gb/s), around one hundred ports and several service classes. Building such a high-speed router would raise many technological problems, one of them being the packet buffer design, mainly because in router design it is important to provide worst-case bandwidth guarantees and not just average-case optimizations. A previous packet buffer design provides worst-case bandwidth guarantees by using a hybrid SRAM/DRAM approach. Next-generation routers need to support hundreds of interfaces (i.e., ports and service classes). Unfortunately, high bandwidth for hundreds of interfaces requires the previous design to use large SRAMs which become a bandwidth bottleneck. The key observation we make is that the SRAM size is proportional to the DRAM access time but we can reduce the effective DRAM access time by overlapping multiple accesses to different banks, allowing us to reduce the SRAM size. The key challenge is that to keep the worst-case bandwidth guarantees, we need to guarantee that there are no bank conflicts while the accesses are in flight. We guarantee bank conflicts by reordering the DRAM requests using a modern issue-queue-like mechanism. Because our design may lead to fragmentation of memory across packet buffer queues, we propose to share the DRAM space among multiple queues by renaming the queue slots. To the best of our knowledge, the design proposed in this paper is the fastest buffer design using commodity DRAM to be published to date.

A DRAM/SRAM memory scheme for fast packet buffers

We address the design of high-speed packet buffers for Internet routers. We use a general DRAM/SRAM architecture for which previous proposals can be seen as particular cases. For this architecture, large SRAMs are needed to sustain high line rates and a large number of interfaces. A novel algorithm for DRAM bank allocation is presented that reduces the SRAM size requirements of previously proposed schemes by almost an order of magnitude, without having memory fragmentation problems. A technological evaluation shows that our design can support thousands of queues for line rates up to 160 Gbps.

A smooth handoff scheme using IEEE802.11 triggers: design and implementation

This paper proposes a handoff scheme in a wireless access network where IEEE802.11 is used as link layer protocol and Mobile IP as network layer protocol. The scheme uses triggers available from IEEE802.11, together with packet buffering in the old Access Point and packet forwarding from the old to the new Access Point in order to provide a smooth handoff. The proposed scheme has been implemented on a Linux based testbed and it has been analysed by means of an ns simulation and an analytical model. The paper reports on the results obtained from the testbed, the simulation and the analytical model, both for constant bit rate traffic (in particular streamed RTP video) and TCP traffic.

A reservation scheme satisfying bandwidth qos constraints for ad-hoc networks

Achieving QoS (Quality of Service) in Mobile Ad-hoc NETworks (MANET) has been a research topic in the last years. In this paper we describe a QoS reservation mechanism for Routing Ad-hoc Networks. The mechanism is targeted for sources requiring a bandwidth allocation. The mechanism is based on the knowledge of the bandwidth requirements of the neighbors of a node and the interferent nodes in the cover area of each node. We describe as the protocol could be integrated in AODV and OLSR. We also give simulation results showing the advantages of our reservation scheme.

Performance Analysis of a Forwarding Scheme for Handoff in HAWAII

Demand for mobile network access is having a huge increase nowadays. Cellular networks are being deployed to cope with a high number of users. Several IP micro-mobility protocols have been proposed to handle routing and handoffs inside cellular networks. In this paper the HAWAII micro-mobility protocol is analyzed by means of an analytical model. A detailed description of the handoff procedure is given and illustrated by means of traces obtained from simulation. Several of the system details are taken into account in the analytical model. This allows us to investigate the influence of various system parameters (e.g. cell overlap area, beacon latency, forwarding buffer capacity, etc.) on the system performance for constant bit rate (UDP) traffic. The results are validated by means of simulation results obtained with the network simulator (ns).

Study of the TCP dynamics over wireless networks with micromobility support using the ns simulator

Several IP micro-mobility protocols have been proposed to enhance the performance of Mobile IP in an environment with frequent handoffs. In this paper we make a detailed study of how some of these protocols namely Cellular IP, HAWAII and Hierarchical Mobile IP affect the behavior of TCP and their interaction with the MAC layer. The aim of the paper is to investigate the impact of handoffs on TCP by means of simulation traces that show the evolution of segments and acknowledgments during handoffs.

A conflict-free memory banking architecture for fast VOQ packet buffers

In order to support the enormous growth of the Internet, innovative research in every router subsystem is needed. We focus our attention on packet buffer design for routers supporting high-speed line rates. More specifically, we address the design of packet buffers using virtual output queuing (VOQ), which are used in most modern router architectures. The design is based on a previously proposed scheme that uses a combination of SRAM and DRAM modules. We propose a storage scheme that achieves a conflict-free memory bank organization. This leads to a reduction of the granularity of DRAM accesses, resulting in a decrease of storage capacity needed by the SRAM. In the DRAM/SRAM scheme, SRAM memory bandwidth needs to fit the line rate. Since memory bandwidth is limited by its size, searching for memory schemes having a small SRAM size arises as an essential issue for high speed line rates (e.g. OC768, 40 Gbps and OC3072, 160 Gbps).

On the design of hybrid DRAM/SRAM memory schemes for fast packet buffers

We address the design of a packet buffer for future high-speed routers that support line rates as high as OC-3072 (160 Gb/s), and a high number of ports and service classes. We describe a general design for hybrid DRAM/SRAM packet buffers that exploits the bank organization of DRAM. This general scheme includes some designs previously proposed as particular cases. Based on this general scheme, we propose a new scheme that randomly chooses a DRAM memory bank for every transfer between SRAM and DRAM. The numerical results show that this scheme would require an SRAM size almost an order of magnitude lower than previously proposed schemes without the problem of memory fragmentation.

A study of the fairness of the fast reservation protocol

Fast Reservation Protocol (FRP) is a Traffic Control scheme intended to multiplex bursty data sources. In this paper we focus on the analysis of the FRP when different sources are multiplexed together in order to study the fairness of the protocol. We present two analytical models to analyse the case in which a set of identical sources is multiplexed with another one of higher rate. Analytical results are compared with simulation results.

A binomial rate control for a media aware Access Point

Multimedia communications over Wireless Local Area Networks like hotspots, are a challenging issue due to the best-effort nature of Internet communications. In a previous work we introduced a Dual-Queue Rate-Controlled mechanism for Access Points (DRAP) in a Hotspot scenario. DRAP has shown to provide some Quality of Service (QoS) guarantees to audio and video traffic flows in the presence of TCP background traffic. In this paper we present an enhancement for DRAP based on a binomial rate control algorithm. We call this enhancement a Binomial DRAP (B-DRAP). Our simulation results show that B-DRAP could be an alternative to provide soft QoS guarantees for multimedia communications.