A Tabu-Based Cache to Improve Range Queries on Prefix Trees

International Journal of Future Computer and Communication, 2013

2008 IEEE International Symposium on Parallel and Distributed Processing, 2008

Hash tables (HTs) are poorly designed for multiple memory accesses during IP lookup and this design flow critically affects their throughput in high-speed routers. Thus, a high capacity HT with a predictable lookup throughput is desirable. A recently proposed fast HT (FHT) [20] has drawbacks like low on-chip memory utilization for a high-speed router and substantial memory overheads due to off-chip duplicate keys and pointers. Similarly, a Bloomier filter-based HT (BFHT) [13], generating an index to a key table, suffers from setup failures and static membership testing for keys. In this paper, we propose a novel hash architecture which addresses these issues by using pipelined Bloom filters. The proposed scheme, a hierarchically indexed HT (HIHT), generates indexes to a key table for the given key, so that the on-chip memory size is reduced and the overhead of pointers in a linked list is removed. Secondly, an HIHT demonstrates approximately 5.1 and 2.3 times improvement in onchip space efficiency with at most one off-chip memory access, compared to an FHT and a BFHT, respectively. In addition to our analyses on access time and memory space, our simulation for IP lookup with 6 BGP tables shows that an HIHT exhibits 4.5 and 2.0 times on-chip memory efficiencies for 160Gbps router than an FHT and a BFHT, respectively.

2010

In this paper, we present and evaluate a protocol that enables fast and accurate range-query execution in Distributed Hash Tables (DHTs). Range queries are of particular importance when the network is populated with groups or collections of data items, whose respective identifiers are generated in a way that encodes semantic relationships into key distances. Contrary to related work in the same direction, our proposed query engine is aware of data replicas at the DHT level and by grouping related nodes into replica neighborhoods, resolves queries with the minimum amount of messaging overhead. Moreover, we suggest pairing respective operations with the core DHT routing mechanics, which allows for reusing existing management and monitoring structures and automatically adapting the query path to the dynamic characteristics of the overlay. We also present an application scenario and the respective deployment details of a prototype implementation in the context of the Gredia project.

Concurrency and Computation: Practice and Experience, 2014

Distributed Hash Tables (DHTs) have been used in several applications, but most DHTs have opted to solve lookups with multiple hops, to minimize bandwidth costs while sacrificing lookup latency. This paper presents D1HT, an original DHT which has a peer-to-peer and self-organizing architecture and maximizes lookup performance with reasonable maintenance traffic, and a Quarantine mechanism to reduce overheads caused by volatile peers. We implemented both D1HT and a prominent single-hop DHT, and we performed an extensive and highly representative DHT experimental comparison, followed by complementary analytical studies. In comparison with current single-hop DHTs, our results showed that D1HT consistently had the lowest bandwidth requirements, with typical reductions of up to one order of magnitude, and that D1HT could be used even in popular Internet applications with millions of users. In addition, we ran the first latency experiments comparing DHTs to directory servers, which revealed that D1HT can achieve latencies equivalent to or better than a directory server, and confirmed its greater scalability properties. Overall, our extensive set of results allowed us to conclude that D1HT can provide a very effective solution for a broad range of environments, from large-scale corporate datacenters to widely deployed Internet applications 1,2 .

2005

Peer-to-peer (P2P) systems provide a robust, scalable and decentralized way to share and publish data. Although highly efficient, current P2P index structures based on Distributed Hash Tables (DHTs) provide only exact match data lookups. This compromises their use in database applications where more advanced query facilities, such as range queries, are a key requirement. In this paper, we give a new P2P indexing structure that supports range searches over shared data while maintaining DHTs logarithmic search time. Our index structure can be seen as an extension of the Chord P2P overlay so that data items are mapped to the Chord address space in an order-preserving way, hence supporting range query executions. Load balancing of skewed data is then achieved deterministically using the underlying DHT infrastructure. Experimental evaluations show that our mechanism provides strong load balancing guarantees in systems with high data skews.

ABSTRACT Massively distributed applications require the integration of heterogeneous data from multiple sources. Peer-to-peer (P2P) is one possible network model for these distributed applications and among P2P architectures, distributed hash table (DHT) is well known for its routing performance guarantees.

IEICE Transactions on Information and Systems, 2011

In this paper, we study the problem of efficient processing of conjunctive queries in Peer-to-Peer systems based on Distributed Hash Tables (P2P DHT, for short). The basic idea of our approach is to cache the search result for the queries submitted in the past, and to use them to improve the performance of succeeding query processing. More concretely, we propose to adopt Bloom filters as a concrete implementation of such a result cache rather than a list of items used in many conventional schemes. By taking such an approach, the cache size for each conjunctive query becomes as small as the size of each file index. The performance of the proposed scheme is evaluated by simulation. The result of simulation indicates that the proposed scheme is particularly effective when the size of available memory in each peer is bounded by a small value, and when the number of peers is 100, it reduces the amount of data transmissions of previous schemes by 75%.

Sigmetrics Performance Evaluation Review, 2011

Over the last decade, storage systems have experienced a 10-fold increase between their capacity and bandwidth. This gap is predicted to grow faster with exponentially growing concurrency levels, with future exascales delivering millions of nodes and billions of threads of execution. A critical component of future file systems for high-end computing is metadata management. This extended abstract presents ZHT, a zero-hop distributed hashtable, which has been tuned for the specific requirements of highend computing. The primary goal of ZHT is excellent availability, fault tolerance, high throughput, and low latencies.

2010

Due to the proliferation of Internet and Intranet, the distributed storage systems have received a lot of attention. These systems span a large number of machines and store huge amount of data for a lot of users. In the distributed storage systems, a row can be directly accessed using a row key. We concentrate on a problem of efficient processing of queries whose predicate is on a column but not a row key. In this paper, we present a cache management technique, called DICE which maintains query results of range queries to support the next range queries. To accelerate the search time of the cached query results, we use modified Interval Ski Lists. In addition, we devise a novel cache replacement policy since DICE maintains an interval rather than a data item. Since our cache replacement policy considers the properties of intervals, our proposed technique is more efficient than traditional buffer replacement algorithms. Our experimental result demonstrates the efficiency of our proposed technique.

2010

Dynamic querying (DQ) is a search technique used in unstructured peer-to-peer (P2P) networks to minimize the number of nodes that is necessary to visit to reach the desired number of results. In this paper, we introduce the use of the DQ technique in structured P2P networks. In particular, we present a P2P search algorithm, named DQ-DHT (Dynamic Querying over a Distributed Hash Table), to perform DQ-like searches over DHT-based overlays.