A Very Efficient Order Preserving Scalable Distributed Data Structure

Abstract

Abstract. SDDSs (Scalable Distributed Data Structures) are access methods specifically designed to satisfy the high performance requirements of a distributed computing environment made up by a collection of computers connected through a high speed network. In this paper we present and discuss performances of ADST, a new order preserving SDDS with a worst-case constant cost for exact-search queries, a worst-case logarithmic cost for update queries, and an optimal worst-case cost for range search queries of O(k) messages, where k is the number of servers covering the query range. Moreover, our structure has an amortized almost constant cost for any single-key query. Finally, our scheme can be easily generalized to manage k-dimensional points, while maintaining the same costs of the 1-dimensional case. We report experimental comparisons between ADST and its direct competitors (i.e., LH*, DRT, and RP*) where it is shown that ADST behaves clearly better. Furthermore we show how our basic technique can be combined with recent proposals for ensuring high-availability to an SDDS. Therefore our solution is very attractive for network servers requiring both a fast response time and a high reliability.

ADST: An Order Preserving Scalable Distributed Data Structure with Constant Access Costs

“…In [7] experimental comparisons exploring behavior of ADST in the average case and comparing it with existing structures are considered. The result is that ADST is the best choice also in the average case.…”

Scalable Distributed Two-Layer Data Structures (SD2DS)

“…Although the size of a record is fixed in some SDDS (like LH*), the rule may be violated to store records of variable size, without breaking the rules of record addressing. A comparison of different SDDS architectures may be found in (Di Pasquale & Nardelli, 2000).…”

Quality of service in wireless networks

“…We relaxed this hypothesis and we provided a preliminary analysis the impact of the topology of the underlying network on the complexity of SDDSs. More precisely, in [21] we defined the problem of allocating servers in a network of workstations. Informally, given a set of already allocated servers Ë in a network , an overloaded server × ¾ Ë , in order to balance its load, may ask for the creation and placement of a new server Ø in a node of not already occupied by another server.…”