Figure 3 - uploaded by Xin Yuan
Content may be subject to copyright.
Source publication
Recent advances in multicasting over the Internet present new opportunities for improving communication performance for clusters of workstations. Realizing collective communication over reliable multicast can achieve higher performance than over reliable unicast primitives. The standard IP multicast, however, only supports unreliable multicast, whi...
Context in source publication
Context 1
... protocols aim at combining the throughput advantage of the NAK{based protocols with the reliability of the ACK{based protocols. Figure 3 shows a ring{based protocol. In ring{based protocols, among all receivers, only one token site is responsible for sending acknowledgment packets to the source. ...
Similar publications
Recent advances in multicasting over the Internet present new opportunities for improving communication performance in clusters of workstations. The standard IP multicast, however only supports unreliable multicast, which is difficult to use for building high level message passing routines. Thus, reliable multicast primitives must be implemented ov...
Recent advances in multicasting over the Internet present new opportunities for improving com-munication performance for clusters of workstations. Realizing collective communication over reli-able multicast can achieve higher performance than over reliable unicast primitives. The standard IP multicast, however, only supports unreliable multicast, w...
Citations
... We now compare our method to related psychoacoustic algorithms solutions [14], [12], [6], [15], [5]. Our algorithm is broadly related to work in the field of operating systems by Maruyama et al., but we view it from a new perspective: client-server algorithms. ...
The growth of users and data throughput of mobile network requires regulation on the data service provider. Thus, the portable device is designed to be accessed to the core network and mirror data for further analysis about the quality of service. Since the portable device is easy to be exposed to shocks while delivered to the work place; and it is cost-dependent for preventive or corrective repair action, in this paper, we conduct a reliability evaluation for the portable Ethernet data acquisition device which is subject to two types of random shocks (extreme shock vs. consecutive-\( k \) minor shock). We define a two-tuple Markov chain to record the occurrence of the two competing failure models and we derive the equation for evaluating the reliability of the device. Furthermore, in order to help the crew to determine whether or not to check the disks after a long time delivery, we, according to the age-based replacement policy, derive the equation for optimizing the disk inspection window by making trade-off between the costs of preventive repair and corrective repair. Finally, we explain the proposed reliability evaluation method with a numerical example.
Cluster platforms have an important role in high performance computing (HPC). They execute cloud computing, data-intensive computing and data center applications, which are supported on distributed file systems. The implementation of data redundancy in these file systems provides a support for high availability and error tolerance. This work proposes an implementation of redundant data storage based on the storage included in the cluster nodes, instead of more expensive approaches with a dedicated storage and network, and on multi-multicast transfers, instead on unicast transfers, to perform the multiple simultaneous data diffusion required for implementing redundant data storage. The proposal applies a recently proposed congestion control scheme that adjusts the sender injection rate, taking into account control information from the receiver nodes and the storage technology available on the cluster nodes. The implementation takes full advantage of the switch diffusion hardware and of the IGMP snooping capability of current switches, which allows to multicast a packet just to the output links with receivers joined to a multicast group. It is made at the user level directly on the UDP interface. Evaluation tests with multiple simultaneous storage accesses were performed in a CentOS cluster. Test results show a more efficient use of the cluster storage. The global bandwidth improves by using hardware related to the storage (network and storage devices) more efficiently.
It is usual that the applications executed in data centers require the distribution of the same data from one node to others at various execution points and that some of them require to cope with multiple of these diffusions in parallel. Multicast-based communications are an alternative solution to sending data efficiently to multiple nodes. This paper proposes a novel technique which offers reliability and congestion control in the multi-multicast transfers in data center networks. The proposal is based on: (1) a new congestion control mechanism, which monitors the control information of the receivers, reducing the server injection rate, (2) taking advantage of the switch diffusion hardware, and (3) using IGMP snooping, which allows a network switch to multicast a packet just to the output links with host receivers joined to a multicast group. The implementation is made at user level and uses the UDP interface. Evaluation tests are performed in a CentOS-based cluster composed of 12 servers in the presence of multiple diffusions at the same time. Test results show improvements in the global bandwidth, avoid network saturation, and reduce overhead included by unicast communications in data transmission.
Multicasting some pieces of information such as messages or packets (called dispatches) from source node(s) to a group of target nodes are governed by a specific sequence in the networked systems. The sequence is called a consensus that indicates an ordering on dispatches to be viewed by the target nodes. Achievement of consensus is a concern in some networked based systems such as distributed ones because the lack of consensus leads to conflict among target nodes reaction. A consensus protocol has some properties to be checked when a source node multicasts a sequence of dispatches to target nodes. The CBCAST protocol is a consensus protocol having properties for ordering and synchronization of dispatches in network communications. This paper thinks of the properties and formulates axioms to check them. The axioms can be practiced for network applications such as group communication and web services. Our approach has two phases consisting of modeling and formulation. The first phase addresses specification of sender and recipient processes by tabular automata. The second phase addresses formulation of axioms using the automaton.
Clusters of workstations employ flexible topologies: regular, irregular, and hierarchical topologies have been used in such
systems. The flexibility poses challenges for developing efficient collective communication algorithms since the network topology
can potentially have a strong impact on the communication performance. In this paper, we consider the all-to-all broadcast
operation on clusters with cut-through and store-and-forward switches. We show that near-optimal all-to-all broadcast on a
cluster with any topology can be achieved by only using the links in a spanning tree of the topology when the message size
is sufficiently large. The result implies that increasing network connectivity beyond the minimum tree connectivity does not
improve the performance of the all-to-all broadcast operation when the most efficient topology specific algorithm is used.
All-to-all broadcast algorithms that achieve near-optimal performance are developed for clusters with cut-through and clusters
with store-and-forward switches. We evaluate the algorithms through experiments and simulations. The empirical results confirm
our theoretical finding.
