ABSTRACT In computer networks with decentralized control, it is necessary to provide a distribute... more ABSTRACT In computer networks with decentralized control, it is necessary to provide a distributed query mechanism, whereby a node can dynamically discover the location of a remote resource. The authors propose a query mechanism based on flooding with feedback, which uses no more than two messages over any link. They examine the fault tolerant aspects of this distributed query mechanism, operating in a network where links and nodes may fail and show that, under certain conditions, the resource is indeed found. They then construct enhancements to this mechanism, where the conditions under which the resource is found are substantially relaxed, at the cost of some increase in message complexity
The objective of this research is how to achieve in engineering practice the potential advantages... more The objective of this research is how to achieve in engineering practice the potential advantages of reliability that a large scale data communication network should make possible. From a technical viewpoint, the principal areas of investigation reported here are computer resource sharing, adaptive network routing, and the efficient representation of network protocol information. The approach has been theoretical, with an
Bibliography: p. 20. "June, 1981." U.S. Dept. of Defense contract No. N00014-75-C-1183 ... more Bibliography: p. 20. "June, 1981." U.S. Dept. of Defense contract No. N00014-75-C-1183 Mario Jodorkovsky and Adrian Segall.
Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc ne... more Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc networks. Initially Bluetooth will be used as a replacement for point-to-(multi)point cables. However, in due course, there will be a need for forming multihop ad-hoc networks over Bluetooth, referred to as scatternets. This paper investigates the capacity assignment problem in Bluetooth scatternets. The problem arises primarily from the special characteristics of the network and its solution requires new protocols. We formulate it as a problem of minimizing a convex function over a convex set contained in the matching polytope. We develop an optimal algorithm which is similar to the well-known flow deviation algorithm and that calls for solving a maximum-weight matching problem at each iteration. A centralized heuristic algorithm with a relatively low complexity is also developed. Then, since in an ad-hoc network there is no central authority that is responsible for network optimization, ...
In this paper we discuss methods for adjusting the layout of virtual paths in an ATM network, to ... more In this paper we discuss methods for adjusting the layout of virtual paths in an ATM network, to the dynamics of changes in the usage of the network by its end-users. We rst present a centralized algorithm for nding a better layout for the current tra c pattern, and for applying the change in the network and discuss its drawbacks. We then present a distributed algorithm that emulates the centralized algorithm with a much lower overhead, and enhanced durability to faults. We prove that both algorithms produce identical layouts, thus showing the superiority of the latter algorithm. Both algorithms base the changes in the network on a new rerouting protocol, which does not cause any losses in data, nor changes in the FIFO order of cells.
This paper describes an architecture for adding anunderlying wavelength-routed optical layer to a... more This paper describes an architecture for adding anunderlying wavelength-routed optical layer to a standardATM network. The optical layer provides reconfigurablelightpaths that serve as links between electronicATM switches. A lightpath is carried on aparticular wavelength and path in the underlying opticalwavelength-routed network. This work focuseson the mechanisms for controlling the optical layer.Distributed control protocols are provided for settingup and taking down
14th International Conference on Distributed Computing Systems, 1994
ABSTRACT A new message-delaying version of synchronizer γ, named ζ, is presented. Synchronizer ζ ... more ABSTRACT A new message-delaying version of synchronizer γ, named ζ, is presented. Synchronizer ζ ensures that original-protocol messages received by a node from nodes in the same cluster are never early, and thus, no buffers for their temporary storage are necessary. Only original-protocol messages on edges leading to nodes of other clusters (external edges) may be early. The z-partition algorithm is introduced to reduce the number of external edges connected to each node, thus reducing the memory overhead of ζ. For an arbitrary z, this algorithm ensures that the external degree of each node is no more than [|V|/z]-1. The z-partition algorithm increases the time complexity of ζ to O(z+logk |V|) per pulse. The tradeoff between memory overhead and time complexity achieved by the z-partition algorithm is optimal
ABSTRACT Synchronizer is the best synchronizer known that works with any type of synchronous mode... more ABSTRACT Synchronizer is the best synchronizer known that works with any type of synchronous model and any network topology. This paper presents three new synchronizers: 1, 2 and . These synchronizers use sparse covers in order to operate and have the following advantages over synchronizer : (1) they are conceptually simpler, as only one convergecast and one broadcast processes are performed along each cluster spanning-tree between each two consecutive pulses, and no preferred links are needed for inter-cluster communication. (2) synchronizer 2 uses half the communication complexity of synchronizer , while retaining the time complexity. (3) synchronizer uses half the time complexity of synchronizer , while retaining the communication complexity. (4) since there is no need to elect preferred links between neighboring clusters, the initialization process of these synchronizers is more efficient: it requires only O(¦V¦log ¦V} + ¦E¦) messages.
ABSTRACT Delaying messages is one of the techniques that can be employed in order to create corre... more ABSTRACT Delaying messages is one of the techniques that can be employed in order to create correct asynchronous protocols using the synchronizer mechanism. In this paper, we show that message delaying cannot be implemented with certain synchronizers and that the synchronizers must be altered before message delaying can be applied. We present three different techniques that solve the problem and work for most synchronizers.
In ATM networks, two levels of connections are defined: virtual path connections and virtual chan... more In ATM networks, two levels of connections are defined: virtual path connections and virtual channel connections. Virtual paths are the building blocks of the virtual channels. This hierarchy, known as “the virtual path concept”, provides considerable advantages in the design of high-speed networks. In the paper the authors show that this concept has the advantage of providing a means of
The terrorist attacks on September 11, 2001 have drawn attention to the use of wireless technolog... more The terrorist attacks on September 11, 2001 have drawn attention to the use of wireless technology in order to locate survivors of structural collapse. We propose to construct an ad hoc network of wireless smart badges in order to acquire information from trapped survivors. We investigate the energy efficient routing problem that arises in such a network and show that
Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc ne... more Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc networks. Initially Bluetooth will be used as a replacement for point-to-(multi)point cables. However, in due course, there will be a need for forming multihop ad-hoc networks over Bluetooth, referred to as scatternets. This paper investigates the capacity assignment problem in Bluetooth scatternets. The problem arises primarily from the special characteristics of the network and its solution requires new protocols. We formulate it as a problem of minimizing a convex function over a polytope contained in the matching polytope. Then, we develop an optimal algorithm which is similar to the well-known flow deviation algorithm and that calls for solving a maximum-weight matching problem at each iteration. Finally, a heuristic algorithm with a relatively low complexity is developed.
ABSTRACT This paper proposes a distributed handover protocol for a micro-cell packet switched mob... more ABSTRACT This paper proposes a distributed handover protocol for a micro-cell packet switched mobile network. In such a network, users move from one cell to another very often, and each change of location may result in misrouted and lost packets. The purpose of the new protocol is to minimize these consequences of location changes: as long as a mobile moves from one cell to another but stays in the same region, the protocol avoids loss of packets and preserves order of transmission. Thus it increases the performance of the transport layer protocol by minimizing the need to retransmit packets
We consider the problem of atomic commitment of a transaction in a distributed database. This is ... more We consider the problem of atomic commitment of a transaction in a distributed database. This is a variant of the famous gossiping problem (see [HID..] for a survey). Given a set of communication costs between pairs of participant sites, we establish that the necessary communication cost for any atomic commitment algorithm is twice the cost of a certain minimwn spanning tree. We also establish ihe necessary communication time for any atomic commiunent algorithm, given a set of communication delays between pairs of participant sites, and the time at which each participant completes its subtransaction. Then we determine that both lower bounds are also upper bounds in the following sense. There is an efficient (i.e. polynomial-time) algorithm that, in the absence of failures, has a minimum communication cost. There is another efficient algorithm that, in the absence of failures, has a minimum communication time. However, unless P=NP, there is no efficient algorithm which has a minimum communication complexity, namely, for which the product of communication cost and communication time is minimwn. Then we present a simple, linear time, distributed algorithm, called TREE-COMMIT, whose communication complexity is not worse than p times the minimum complexity, where p is the number of participants. Finally. we demonstrate that TREE-COMMIT is superior to the existing variants of the two-phase commit protocol.
ABSTRACT In computer networks with decentralized control, it is necessary to provide a distribute... more ABSTRACT In computer networks with decentralized control, it is necessary to provide a distributed query mechanism, whereby a node can dynamically discover the location of a remote resource. The authors propose a query mechanism based on flooding with feedback, which uses no more than two messages over any link. They examine the fault tolerant aspects of this distributed query mechanism, operating in a network where links and nodes may fail and show that, under certain conditions, the resource is indeed found. They then construct enhancements to this mechanism, where the conditions under which the resource is found are substantially relaxed, at the cost of some increase in message complexity
The objective of this research is how to achieve in engineering practice the potential advantages... more The objective of this research is how to achieve in engineering practice the potential advantages of reliability that a large scale data communication network should make possible. From a technical viewpoint, the principal areas of investigation reported here are computer resource sharing, adaptive network routing, and the efficient representation of network protocol information. The approach has been theoretical, with an
Bibliography: p. 20. "June, 1981." U.S. Dept. of Defense contract No. N00014-75-C-1183 ... more Bibliography: p. 20. "June, 1981." U.S. Dept. of Defense contract No. N00014-75-C-1183 Mario Jodorkovsky and Adrian Segall.
Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc ne... more Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc networks. Initially Bluetooth will be used as a replacement for point-to-(multi)point cables. However, in due course, there will be a need for forming multihop ad-hoc networks over Bluetooth, referred to as scatternets. This paper investigates the capacity assignment problem in Bluetooth scatternets. The problem arises primarily from the special characteristics of the network and its solution requires new protocols. We formulate it as a problem of minimizing a convex function over a convex set contained in the matching polytope. We develop an optimal algorithm which is similar to the well-known flow deviation algorithm and that calls for solving a maximum-weight matching problem at each iteration. A centralized heuristic algorithm with a relatively low complexity is also developed. Then, since in an ad-hoc network there is no central authority that is responsible for network optimization, ...
In this paper we discuss methods for adjusting the layout of virtual paths in an ATM network, to ... more In this paper we discuss methods for adjusting the layout of virtual paths in an ATM network, to the dynamics of changes in the usage of the network by its end-users. We rst present a centralized algorithm for nding a better layout for the current tra c pattern, and for applying the change in the network and discuss its drawbacks. We then present a distributed algorithm that emulates the centralized algorithm with a much lower overhead, and enhanced durability to faults. We prove that both algorithms produce identical layouts, thus showing the superiority of the latter algorithm. Both algorithms base the changes in the network on a new rerouting protocol, which does not cause any losses in data, nor changes in the FIFO order of cells.
This paper describes an architecture for adding anunderlying wavelength-routed optical layer to a... more This paper describes an architecture for adding anunderlying wavelength-routed optical layer to a standardATM network. The optical layer provides reconfigurablelightpaths that serve as links between electronicATM switches. A lightpath is carried on aparticular wavelength and path in the underlying opticalwavelength-routed network. This work focuseson the mechanisms for controlling the optical layer.Distributed control protocols are provided for settingup and taking down
14th International Conference on Distributed Computing Systems, 1994
ABSTRACT A new message-delaying version of synchronizer γ, named ζ, is presented. Synchronizer ζ ... more ABSTRACT A new message-delaying version of synchronizer γ, named ζ, is presented. Synchronizer ζ ensures that original-protocol messages received by a node from nodes in the same cluster are never early, and thus, no buffers for their temporary storage are necessary. Only original-protocol messages on edges leading to nodes of other clusters (external edges) may be early. The z-partition algorithm is introduced to reduce the number of external edges connected to each node, thus reducing the memory overhead of ζ. For an arbitrary z, this algorithm ensures that the external degree of each node is no more than [|V|/z]-1. The z-partition algorithm increases the time complexity of ζ to O(z+logk |V|) per pulse. The tradeoff between memory overhead and time complexity achieved by the z-partition algorithm is optimal
ABSTRACT Synchronizer is the best synchronizer known that works with any type of synchronous mode... more ABSTRACT Synchronizer is the best synchronizer known that works with any type of synchronous model and any network topology. This paper presents three new synchronizers: 1, 2 and . These synchronizers use sparse covers in order to operate and have the following advantages over synchronizer : (1) they are conceptually simpler, as only one convergecast and one broadcast processes are performed along each cluster spanning-tree between each two consecutive pulses, and no preferred links are needed for inter-cluster communication. (2) synchronizer 2 uses half the communication complexity of synchronizer , while retaining the time complexity. (3) synchronizer uses half the time complexity of synchronizer , while retaining the communication complexity. (4) since there is no need to elect preferred links between neighboring clusters, the initialization process of these synchronizers is more efficient: it requires only O(¦V¦log ¦V} + ¦E¦) messages.
ABSTRACT Delaying messages is one of the techniques that can be employed in order to create corre... more ABSTRACT Delaying messages is one of the techniques that can be employed in order to create correct asynchronous protocols using the synchronizer mechanism. In this paper, we show that message delaying cannot be implemented with certain synchronizers and that the synchronizers must be altered before message delaying can be applied. We present three different techniques that solve the problem and work for most synchronizers.
In ATM networks, two levels of connections are defined: virtual path connections and virtual chan... more In ATM networks, two levels of connections are defined: virtual path connections and virtual channel connections. Virtual paths are the building blocks of the virtual channels. This hierarchy, known as “the virtual path concept”, provides considerable advantages in the design of high-speed networks. In the paper the authors show that this concept has the advantage of providing a means of
The terrorist attacks on September 11, 2001 have drawn attention to the use of wireless technolog... more The terrorist attacks on September 11, 2001 have drawn attention to the use of wireless technology in order to locate survivors of structural collapse. We propose to construct an ad hoc network of wireless smart badges in order to acquire information from trapped survivors. We investigate the energy efficient routing problem that arises in such a network and show that
Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc ne... more Bluetooth enables portable electronic devices to communicate wirelessly via short-range ad-hoc networks. Initially Bluetooth will be used as a replacement for point-to-(multi)point cables. However, in due course, there will be a need for forming multihop ad-hoc networks over Bluetooth, referred to as scatternets. This paper investigates the capacity assignment problem in Bluetooth scatternets. The problem arises primarily from the special characteristics of the network and its solution requires new protocols. We formulate it as a problem of minimizing a convex function over a polytope contained in the matching polytope. Then, we develop an optimal algorithm which is similar to the well-known flow deviation algorithm and that calls for solving a maximum-weight matching problem at each iteration. Finally, a heuristic algorithm with a relatively low complexity is developed.
ABSTRACT This paper proposes a distributed handover protocol for a micro-cell packet switched mob... more ABSTRACT This paper proposes a distributed handover protocol for a micro-cell packet switched mobile network. In such a network, users move from one cell to another very often, and each change of location may result in misrouted and lost packets. The purpose of the new protocol is to minimize these consequences of location changes: as long as a mobile moves from one cell to another but stays in the same region, the protocol avoids loss of packets and preserves order of transmission. Thus it increases the performance of the transport layer protocol by minimizing the need to retransmit packets
We consider the problem of atomic commitment of a transaction in a distributed database. This is ... more We consider the problem of atomic commitment of a transaction in a distributed database. This is a variant of the famous gossiping problem (see [HID..] for a survey). Given a set of communication costs between pairs of participant sites, we establish that the necessary communication cost for any atomic commitment algorithm is twice the cost of a certain minimwn spanning tree. We also establish ihe necessary communication time for any atomic commiunent algorithm, given a set of communication delays between pairs of participant sites, and the time at which each participant completes its subtransaction. Then we determine that both lower bounds are also upper bounds in the following sense. There is an efficient (i.e. polynomial-time) algorithm that, in the absence of failures, has a minimum communication cost. There is another efficient algorithm that, in the absence of failures, has a minimum communication time. However, unless P=NP, there is no efficient algorithm which has a minimum communication complexity, namely, for which the product of communication cost and communication time is minimwn. Then we present a simple, linear time, distributed algorithm, called TREE-COMMIT, whose communication complexity is not worse than p times the minimum complexity, where p is the number of participants. Finally. we demonstrate that TREE-COMMIT is superior to the existing variants of the two-phase commit protocol.
Uploads
Papers by Adrian Segall