SUMMARY: Cancer hallmarks rely on its specific transcriptional programs, which are dysregulated by multiple mechanisms, including genomic aberrations in the DNA regulatory regions. Genome-wide association studies have shown many variants are found within putative enhancer elements. To provide insights into the regulatory role of enhancer-associated non-coding variants in cancer epigenome, and to facilitate the identification of functional non-coding mutations, we present dbInDel, a database where we have comprehensively analyzed enhancer-associated insertion and deletion variants for both human and murine samples using ChIP-Seq data. Moreover, we provide the identification and visualization of upstream TF binding motifs in InDel-containing enhancers. Downstream target genes are also predicted and analyzed in the context of cancer biology. The dbInDel database promotes the investigation of functional contributions of non-coding variants in cancer epigenome. AVAILABILITY AND IMPLEMENTATION: The database, dbInDel, can be accessed from http://enhancer-indel.cam-su.org/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

TINA (Telecommunications Information Networking Architecture) is a world-wide initiative with the objective of defining an architecture for the future telecommunications infrastructure, which should be able to handle multimedia services. This paper investigates and discusses the concepts currently available in the TINA architecture to model multimedia services. The focus of the paper is targeted towards the support for describing, requiring and setting up the necessary multimedia connections between the parties involved in distributed multimedia services. By means of an example, we illustrate how the TINA modeling concepts can be successfully applied to the description of a multimedia conferencing service scenario.

Ethernet at 10 Mb/s and other legacy LANs such as 4 and 16 Mb/s Token Ring have matured and proliferated widely over the last decade. Now, applications have started to exceed the capabilities of these low speed LANs. A new generation of high speed networks are being deployed to meet with increasing expectations of users. This paper compares four such high speed LANs which operate at or above 100 Mb/s: 100BASE-T, 100VG-AnyLAN, FDDI and ATM. The attributes and performance of these four LANs are described and contrasted. The implications of migration to these new technologies are discussed.>

We analyze the behavior of the Ethernet in networks with a small number of active stations, and describe what is known as the Ethernet capture effect. Where a station transmits consecutive packets exclusively for a prolonged period despite other stations contending for access. The capture effect causes transient unfairness, which results in substantial performance degradation. We report measurements using TCP/IP that show the performance degradation. A solution is proposed that effectively overcomes the capture effect. The proposed algorithm, which we call the capture avoidance binary exponential backoff (CABEB), uses the standard binary exponential backoff (BEB) with enhancements for collision resolution in the special case when a station attempts to capture the channel subsequent to an uninterrupted consecutive transmit. Using a detailed simulation, we show the efficacy of the CABEB algorithm over the standard BEB in overcoming the unfairness resulting from stations capturing the channel. The CABEB improves throughput for protocols like TCP/IP, reduces variability in the channel access latency and eliminates packet discards due to excessive collisions in a 2-node network. The algorithm is a modification that is compliant with the Ethernet/802.3 standards. For networks with a large number of active stations, the CABEB performs as well as the standard BEB algorithm. Our study placer emphasis on the workload and network configuration that is the worst case relative to the Ethernet capture effect to show that the proposed algorithm is a substantial improvement over the existing backoff algorithm.>

We address issues in determining FIFO sizes necessary for high-performance, in an integrated high-speed network interface, using a 100 Mbps Fast Ethernet controller as an example. A detailed analytical model is developed which accounts for system design choices, in addition to network parameters such as packet size and rate. The model yields insight into the impact of system parameters, such memory latency and maximum DMA transfer size, on the size of FIFOs required. The model also shows that the worst-case, in terms of receive-FIFO required, is not necessarily when back-to-back minimum size packets are received, but depends on the system parameters such as maximum DMA transfer size. We also study the possibility of FIFO underflows for the transmit direction.>

Duplicate station addresses can disrupt operation of a token ring. Failure symptoms include failure to initialize the ring (make a token), false re-initialization of the ring, and the inability of certain stations to communicate with other stations. These problems were recognized and some solutions for simple cases are included in the FDDI standards Additional cases result from common usage of station addresses which are not solved by the standard algorithms. For example, stations may have multiple individual addresses and may use some addresses as destination addresses but not source addresses. These more complicated fault cases and solutions for them are discussed.>

The fiber distributed data interface (FDDI) is a 100 Mb/s token ring which includes extensive reliability and robustness mechanisms to provide fault detection, isolation, monitoring, and recovery functions. The authors examine two well-known problems on token rings and their potential impact on FDDI. The two problems are no-owner frame (NOF) and duplicate tokens. FDDI effectively resolves these problems only when either the NOF or the duplicate token collides into a transmitting station. The impact of NOFs and duplicate tokens during the period from the occurrence of the NOF and the duplicate token until they are removed by a transmitting station can be severe. This paper presents a ring purger that transparently and continuously removes the NOFs and detects and removes duplicate tokens. A ring purger is a designated station on a ring that performs its normal functions as a station and, in addition, performs the removal of NOFs and duplicate tokens. A distributed election algorithm is used to elect a single station on a given ring to be the ring purger. The authors also present performance and robustness considerations used in the design of the ring purger algorithms. They study the performance effect of continuously purging the ring and show that the effect is negligible even in the worst case.>

A frame content independent stripping (FCIS) algorithm for token rings is described that can be used by source-routing bridges due to its simplicity and the fact that the resulting size of the fragments remaining on the ring is very small. The FCIS algorithm counts the number of outstanding frames that a station has transmitted but not yet stripped from the ring and in addition places a global delimiter such as a void frame at the end of the transmission of the information frames. The station then stops stripping when either the count goes to zero or when the void frame or token is received by the station. The algorithm is simple to implement and does not place any topological restrictions on the network. It also allows for interoperation with stations which do not implement the FCIS algorithm. With the FCIS algorithm, frames are stripped at the earliest possible moment. It is shown that the combination of both the count and void is necessary for an FCIS algorithm by demonstrating that the inclusion of either just the void or the count could result in considerable overstripping of frames on the ring due to errors on the ring. The void frame adds a total of 28 bytes of overhead to every transmission opportunity at a participating station.>

Token rings have the property that a station that transmits a frame on the ring is responsible for removing the frame after it has been delivered to the destination stations. The algorithm to perform the frame removal is called 'Frame Stripping'. Most existing algorithms strip frames based on their content. This is not always adequate. The need for a new algorithm arises from the fact that frames transmitted by a station need not have the station's own address as the source address for a variety of reasons - such as when a bridge transmits a frame or when another address is used as the source address by a station instead of its original station address. This paper discusses a new frame content independent stripping (FCIS) algorithm for token rings.