Site selection for real-time client request handling

Kanitkar, Vinay; Delis, Alex

doi:10.1109/icdcs.1999.776531

Cited by 3 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is because transaction response times are mostly determined by database related factors like disk I/O, data contention due to concurrent access, and bookkeeping overhead of database recovery than CPU waits [2]. Furthermore, load balancing algorithms in distributed systems often migrate a task to other node during its execution or decompose a task into many concurrent subtasks [8,9,15]. We will not consider transaction migration in this paper.…”

Section: Related Workmentioning

confidence: 99%

Dynamic Affinity Cluster Allocation in a Shared Disks Cluster

Ohn

Cho

2006

J Supercomput

View full text Add to dashboard Cite

A shared disks (SD) cluster couples multiple computing nodes for high performance transaction processing, and all nodes share a common database at the disk level. In the SD cluster, a front-end router selects a node for an incoming transaction to be executed. An affinity-based routing can increase the buffer hit ratio of each node by clustering transactions referencing similar data to be executed on the same node. However, the affinity-based routing is non-adaptive to the changes of the system load. This means that a specific node would be overloaded if corresponding transactions rush into the system. In this paper, we propose a new transaction routing algorithm, named Dynamic Affinity Cluster Allocation (DACA). DACA can make an optimal balance between the affinity-based routing and indiscriminate sharing of load in the SD cluster. As a result, DACA can increase the buffer hit ratio and reduce the frequency of inter-node buffer invalidations while achieving the dynamic load balancing.

show abstract

Section: Related Workmentioning

confidence: 99%

Dynamic Affinity Cluster Allocation in a Shared Disks Cluster

Ohn

Cho

2006

J Supercomput

View full text Add to dashboard Cite

show abstract

“…The fundamental idea in NOW database systems is to carry out simultaneous I/O operations whenever possible and to execute highly intensive CPU processing in a distributed fashion [43,66]. Computationally intensive problems such as the N -body problem have been addressed in the context of NOW environments.…”

Section: Related Workmentioning

confidence: 99%

Radio-wave propagation prediction using ray-tracing techniques on a network of workstations (NOW)

Chen¹,

Delis

Bertoni³

2004

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Ray-tracing based radio wave propagation prediction models play an important role in the design of contemporary wireless networks as they may now take into account diverse physical phenomena including reflections, diffractions, and diffuse scattering. However, such models are computationally expensive even for moderately complex geographic environments. In this paper, we propose a computational framework that functions on a network of workstations (NOW) and helps speed up the lengthy prediction process. In ray-tracing based radio propagation prediction models, orders of diffractions are usually processed in a stage-by-stage fashion. In addition, various source points (transmitters, diffraction corners, or diffuse scattering points) and different ray-paths require different processing times. To address these widely varying needs, we propose a combination of the phase-parallel and manager/workers paradigms as the underpinning framework. The phase-parallel component is used to coordinate different computation stages, while the manager/workers paradigm is used to balance workloads among nodes within each stage. The original computation is partitioned into multiple small tasks based on either raypath-level or source-point-level granularity. Dynamic load-balancing scheduling schemes are employed to allocate the resulting tasks to the workers.We also address issues regarding main memory consumption, intermediate data assembly, and final prediction generation. We implement our proposed computational model on a NOW configuration by using the message passing interface (MPI) standard. Our experiments with real and synthetic building and terrain databases show that, when no constraint is imposed on the main memory consumption, the proposed prediction model performs very well and achieves nearly linear speedups under various workload. When main memory consumption is a concern, our model still delivers very promising performance rates provided that the complexity of the involved computation is high, so that the extra computation and communication overhead introduced by the proposed model do not dominate the original computation. The accuracy of prediction results and the achievable speedup rates can be significantly improved when 3D building and terrain databases are used and/or diffuse scattering effect is taken into account.Indexing Terms: Ray-tracing based radio propagation prediction model, Radio wave prediction model on Network of Workstations (NOW), Data or control domain decomposition, Dynamic workload-balancing scheduling schemes.

show abstract

“…This allows the system to continue to make progress on late transactions, but only when no transactions with earlier deadlines are available for consideration. In our previous work [24,25], we have shown that a client-server database system can be more efficient than a centralized system (for real-time processing) in the presence of the following conditions: (i) if there is a reasonable amount of spatial and temporal locality in client data access patterns, and (ii) the percentage of data modifications is low. However, the presence of a high percentage of updates can cause centralized systems to perform better than their client-server counterparts.…”

Section: Efficient Access For Massive E-commerce User Communitiesmentioning

confidence: 99%

Efficient Processing of Client Transactions in Real-Time

Kanitkar

Delis²

2005

Distributed and Parallel Databases

View full text Add to dashboard Cite

Abstract. In traditional client-server databases, a transaction and its requisite data have to be colocated at a single site for the operation to proceed. This has usually been achieved by moving either the data or the transaction. However, the availability of high-bandwidth networking options has led users of today's systems to expect realtime guarantees about the completion time of their tasks. In order to offer such guarantees in a client-server database system, a transaction should be processed by any means that allows it to meet its deadline. To this end, we explore the option of moving both transactions and data to the most promising sites for successful completion. We propose a load-sharing framework that oversees the shipment of data and transactions so as to increase the efficiency of a cluster consisting of a server and a number of clients. Here, efficiency is defined as the percentage of transactions successfully completed within their deadlines by the cluster. The suitability of a client for processing a transaction is measured with respect to the availability of the transaction's required data in its local cache. In addition to the load-sharing algorithm, we use the concept of grouped locks, along with transaction deadline information, in order to schedule the movement of data objects in the cluster in a more efficient manner. We evaluate the real-time processing performance of the client-server architecture using detailed experimental testbeds. Our evaluation indicates that it is possible, in many situations, to achieve better performance than a centralized system.

show abstract

Site selection for real-time client request handling

Cited by 3 publications

References 23 publications

Dynamic Affinity Cluster Allocation in a Shared Disks Cluster

Dynamic Affinity Cluster Allocation in a Shared Disks Cluster

Radio-wave propagation prediction using ray-tracing techniques on a network of workstations (NOW)

Efficient Processing of Client Transactions in Real-Time

Contact Info

Product

Resources

About