A. Trachtenberg, D. Starobinski, and S. Agarwal, "Fast PDA synchronization using characteristic polynomial interpolation," Proc. INFOCOM, June 2002.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....we can then synchronize them using either the remote file synchronization techniques presented earlier, or by transmitting the entire item in the case of small database records. In the following, we discuss the problems arising in this scenario, with emphasis on a recent approach described in [26, 33, 46]. Some earlier work on reconciliation of record based data appeared in [7, 1, 31, 32] Consider the case, assumed in [46] of a handheld device using the Palm Hotsync program to synchronize its database of addresses or appointments with a desktop device. If the handheld was last synchronized with ....

....entire item in the case of small database records. In the following, we discuss the problems arising in this scenario, with emphasis on a recent approach described in [26, 33, 46] Some earlier work on reconciliation of record based data appeared in [7, 1, 31, 32] Consider the case, assumed in [46], of a handheld device using the Palm Hotsync program to synchronize its database of addresses or appointments with a desktop device. If the handheld was last synchronized with the same desktop, then the Palm Hotsync software can use auxiliary logging information to efficiently identify items that ....

[Article contains additional citation context not shown here]

A. Trachtenberg, D. Starobinski, and S. Agarwal. Fast PDA synchronization using characteristic polynomial interpolation. Technical Report BU-2001-03, Boston University, Department of Electrical and Computer Engineering, 2001.

No context found.

A. Trachtenberg, D. Starobinski, and S. Agarwal, "Fast PDA synchronization using characteristic polynomial interpolation," Proc. INFOCOM, June 2002.

....under grants MIP9630096 and CCR 0133521. 1 Introduction The problem of reconciling data is inherent to applications that require consistency among distributed information, including diverse examples such as gossip protocols for distributing networked data [1] resource discovery [2] mobile data [3, 4], and sequences of symbols from a given alphabet, such as nucleotide sequences in DNA or amino acids sequences in proteins [5] In each of these examples, the system needs to determine and, thereafter, reconcile the differences between data stored in physically separate locations. From the ....

....(in) 5.2 Classical reconciliation Consider the traditional formulation of the set reconciliation problem, as stated in Section 2.3. This form of set reconciliation has many applications in networking, including gossip protocols [1, 21, 22] resource discovery [2, 6] and synchronization [3, 4]. In the context of this work, we may view classical set reconciliation as a one way data reconciliation problem under Hamming errors of weight at most t. Corollary 3 shows that the minimum number of bits needed for such reconciliation when t = n is log2(maxx I(x)l) 2 b. In other words, the ....

A. Trachtenberg, D. Starobinski, and S. Agarwal, "Fast PDA synchronization using char- acteristic polynomial interpolation," Proc. INFOCOM, June 2002.

....depends only on the data set size, whereas for CPIsync the regression is obtained by fitting the data to a cubic polynomial that depends only on the number of differences. With such analytical models, one can determine a threshold for any given set size and number of differences between hosts [28]. In a Palm PDA application like an address book or memo, changes between concurrent synchronizations typically involve only a small number of records, in which case CPIsync will usually be much faster than slow sync. Predictably, slow sync also expends much more energy to synchronize than ....

A. Trachtenberg, D. Starobinski, and S. Agarwal, "Fast PDA synchronization using characteristic polynomial interpolation," Proc. INFOCOM, June 2002. 28

.... however, is that the interpolation and factorization steps demand a cubic computation time (in the number of differences) The impact of this is lessened by the fact that the computation can be distributed asymmetrically according to the relative computing power of the synchronizing machines [1, 4]. We have implemented and deployed the CPISync algorithm in a PC PalmPilot synchronization system, which we feel accurately models a peer to peer link in a heterogeneous network. Our PDA application, a memo entry program called SyncMemo, uses CPISync to synchronize data with its PC counterpart ....

A. Trachtenberg, D. Starobinski, and S. Agarwal, Fast PDA Synchronization using Characteristic Polynomial Interpolation, Proceedings of the IEEE INFOCOM 2002.

....of bandwidth, processing, and memory resources. In particular, we show that existing synchronization protocols often do not scale well with the number of devices in a network, and with the size of the data to be synchronized. We compare these protocols to a new algorithm, called CPISync [2], which addresses some of the scalability issues. Our conclusions are intended to provide guidance for addressing the scalability issues inherent to synchronizing data over large, heterogeneous, tetherless networks. This paper is organized as follows. In Section 2, we describe some fundamentals ....

....for mobile devices. In particular we explain Palm s Hot sync synchronization protocol [3] Pumatech s Intellisync [4] synchronization protocol, and a recent industry wide initiative called the SyncML [5] standard. This section also includes a description of the new scalability oriented CPISync [2] protocol. In Section 4 we provide a taxonomy of the reviewed protocols according to various scalability metrics, and consider future trends. Section 5 is devoted to concluding remarks and directions for further research. 2 Fundamentals of PDA Synchronization A PDA can be represented by a node ....

[Article contains additional citation context not shown here]

A. Trachtenberg, D. Starobinski, and S. Agarwal. Fast PDA synchronization using characteristic polynomial interpolation. In IEEE Infocom, 2002. To appear.

....of bandwidth, processing, and memory resources. In particular, we show that existing synchronization protocols often do not scale well with the number of devices in a network, and with the size of the data to be synchronized. We compare these protocols to a new algorithm, called CPISync [2], which addresses some of the scalability issues. Our conclusions are intended to provide guidance for addressing the scalability issues inherent to synchronizing data over large, heterogeneous, tetherless networks. This paper is organized as follows. In Section 2, we describe some fundamentals ....

....for mobile devices. In particular we explain Palm s Hot sync synchronization protocol [3] Pumatech s Intellisync [4] synchronization protocol, and a recent industry wide initiative called the SyncML [5] standard. This section also includes a description of the new scalability oriented CPISync [2] protocol. In Section 4 we provide a taxonomy of the reviewed protocols according to various scalability metrics, and consider future trends. Section 5 is devoted to concluding remarks and directions for further research. 2 Fundamentals of PDA Synchronization A PDA can be represented by a node ....

[Article contains additional citation context not shown here]

A. Trachtenberg, D. Starobinski, and S. Agarwal. Fast PDA synchronization using characteristic polynomial interpolation. In IEEE Infocom, 2002. To appear.

....SA [ SB with a minimum of communication. Set reconciliation is connected to graph coloring and the construction of error correcting codes [6, 7] and has a number of applications outside of gossip protocols. In particular, it has been applied to PDA synchronization and routing table maintenance [8, 13, 14]. More generally, set reconciliation can be used in a variety of systems where distributed information needs to be reconciled. Earlier work in [10, 11] presented algorithms for solving set reconciliation with near optimal communication complexity by interpolating rational functions over a nite ....

....an error probability of k , as de ned below: jS A j jS B j 2 b k ; 3) The general framework for the algorithms in [10, 11] collectively called Basic Recon) can be abstracted into four primitives, shown in Figure 1. The details of how these primitives are implemented can be found in [10, 11, 14]. The general idea is that the checksum of a set is computed 2 init cs(m; k) Returns an initial checksum with recovery bound m and redundancy factor k; the size of the checksum is given in Equation (4) add element(cs; Updates cs to re ect addition of to host s set. del element(cs; ....

Trachtenberg, A., Starobinski, D., and Agarwal, S. Fast PDA synchronization using characteristic polynomial interpolation. Proc. INFOCOM (June 2002). to appear.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC