Showing papers by "Yehuda Afek published in 2000"

Long-lived and adaptive atomic snapshot and immediate snapshot (extended abstract)

Abstract: Long-lived and adaptive to point contention implementations of snapshot and immediate snapshot objects in the read/write shared-memory model are presented. In [2] we presented adaptive algorithms for mutual exclusion, collect and snapshot. However, the collect and snapshot algorithms were adaptive only when the number of local primitive operations that a process performs are ignored, i.e., not counted. The number of primitive local steps (operations that do not access the shared memory) in the collect and snapshot operations presented in [2] is O(Nk3) and O(Nk4) respectively where N is the total number of processes in the system and k is the encountered contention. Here we developed new techniques that enabled us to achieve fully adaptive implementations in which the step complexity (combined local and shared) of any operation is bounded by a function of the number of processes that are concurrent with the operation, in particular, O(k4) for the snapshot implementation.

Bounds on the shared memory requirements for long-lived & adaptive objects (extended abstract)

Abstract: In this paper we prove: For any constant d there is a large enough n such that there is no long-lived adaptive implementation of collect or renaming in the read write model with n processes that uses d or less MWMR registers.In other words, there is no implementation of a long-lived and adaptive renaming or collect object in the atomic read/write model that uses O(1) multi-writer-multi-reader registers and any number of single-writer-multi-reader registers. In 1980 Burns and Lynch [1] proved that at least n multi-writer-multi-reader (MWMR) registers are necessary in any mutual exclusion algorithm that uses only MWMR registers (i.e., atomic registers). It is also relatively easy to see that any adaptive non-trivial algorithm uses at least one multi-writer-multi-reader (MWMR) register even when there are n single-writer-multi-reader (SWMR) registers. Here we extend the techniques of Burns and Lynch and prove that adaptive algorithms that use both SWMR and MWMR registers such as, collect and renaming, need in addition to the O(n) SWMR registers a non-constant, F(n) number of MWMR registers.

Method for routing with a clue

Abstract: A method of routing a data packet from a forwarding router to a downstream router. The data packet header includes an address that includes a bit string. The forwarding router looks up, in a forwarding database, a prefix that best matches the bit string. The forwarding router then attaches to the data packet a clue that is related to the best matching prefix, and forwards the data packet to the downstream router. The downstream router looks up, in a downstream database, and with reference to the clue, another prefix that best matches the bit string. Because the databases of neighboring routers are similar, the clue either directly determines the best matching prefix at the downstream router or provides the downstream router with a good starting point for its lookup.

Method for fast multi-dimensional packet classification

Abstract: A simple and fast algorithm for multi-dimensional packet classification by solving the best matching filter problem. The substantial part of the algorithm includes the search of filters being concurrently stabbed by the packet using a KD-tree data structure. Another aspect of the present invention includes the classification of the packet according to its destination address using a second data-structure which preferably consists of a one dimensional segment tree. In a preferred embodiment of the present invention, the packet is first classified according to its protocol type, then the packet is classified according to its destination address using a one dimensional segment tree data structure, and finally, a 6 dimension KD-tree is used to find the filters being stabbed by the remaining 3 parameters of the packet. Among the filters, which comply with the packet, the filter which applies to the packet is the filter with the highest pre-determined priority. Theoretical analysis predicts and experimental results verify that in practice, the proposed data-structure can handle a lot of “difficult” filters quickly and efficiently.

Trainet: a new label switching scheme

Abstract: Trainet, a new scheme to extend MPLS (multi-protocol label switching) is presented. The scheme works much like the subway system in a large metropolitan area. Each (unidirectional) subway line corresponds to a labeled path, and a route in the network is defined by either a pair , where count specifies how many hops a packet still has to take in the specified train, or a route may be defined by a sequence of such pairs. A sequence of such pairs specifies that the packet has to take a number of hops in one train-line, and then continue for a certain number of hops on another train-line and so forth. While slightly increasing the number of labels in a header and adding a counter to each label, the scheme reduces the total number of different labels necessary in the network, and in each switch. Thus, for a given number of labels it may support a larger number of flows. Moreover, our scheme considerably simplifies the path set-up cost while still providing all the features of MPLS: switching, supporting QoS, explicit routing, and traffic engineering.