QoS control method to reduce resource reservation failure in datagrid applications

Noro, Masaaki; K, Baba; Shimojo, Shinsuke

doi:10.1109/pacrim.2005.1517330

Cited by 7 publications

(7 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method is a combination of a dynamic reserved bandwidth control [13] and data transfer using two kinds of TCP. It [13] transfers data using a single TCP NewReno connection and reserves network bandwidth.…”

Section: Our Methodsmentioning

confidence: 99%

“…It [13] transfers data using a single TCP NewReno connection and reserves network bandwidth. The method proposed here uses two kinds of TCP connections for one data transfer.…”

Section: Our Methodsmentioning

confidence: 99%

“…The dynamic reservation bandwidth control method and the combination of two methods are denoted in the figure by DynQoS and DynQoS+TwoTCP, respectively. The "DynQoS" method is the same as the method reported by [13]. It is difficult to compare the performance of the different meth- ods when there is call loss, which is the termination of data transfers that fail to reserve network bandwidth.…”

Section: ) Deadline Of Data Transfermentioning

confidence: 99%

See 2 more Smart Citations

A large-scale data transfer method that reduces reservation bandwidth over DiffServ AF networks

Noro

Shimojo

2008

2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks

View full text Add to dashboard Cite

We have developed a data transfer method that uses less reserved bandwidth on DiffServ AF networks. Our method consists of a dynamic QoS control method and a data transfer protocol that uses two kinds of TCP connections. We simulated the transfer of data using our method on a network simulator and found that it performed better than current methods. By reducing reserved bandwidth, our method will reduce the cost to network providers of meeting service level agreements with customers. It will be particularly useful in such applications as pipeline processing in large-scale grid computing and remote data backup over wide area networks because these applications will need to reserve less bandwidth than they currently need to meet job schedules.

show abstract

Section: Our Methodsmentioning

confidence: 99%

“…It [13] transfers data using a single TCP NewReno connection and reserves network bandwidth. The method proposed here uses two kinds of TCP connections for one data transfer.…”

Section: Our Methodsmentioning

confidence: 99%

Section: ) Deadline Of Data Transfermentioning

confidence: 99%

See 1 more Smart Citation

A large-scale data transfer method that reduces reservation bandwidth over DiffServ AF networks

Noro

Shimojo

2008

2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks

View full text Add to dashboard Cite

show abstract

“…When used to a disc-todisc transfer, the throughput reached was 58.3% in the same link [33]. In conclusion, we assumed that our Grid traffic is composed of GridFTP sessions that arrive following a stationary Poisson process with several intensities according to the average arrival rates limits in [23]. Also, we assume that job sizes follow an exponential distribution with means 100MB or 500MB.…”

Section: Grid Traffic Modelmentioning

confidence: 99%

“…We checked that the first 5 minutes of each simulation run correspond to the transient period for reaching the equilibrium. Arrival intensities were taken from [23] and ranges from [0,20] arrivals per minute of GridFTP sessions. For each scenario, thirty replications were carried out.…”

Section: Simulation Experimentsmentioning

confidence: 99%

Performance comparison of Flow Aware Networking (FAN) architectures under GridFTP traffic

Cárdenas

Gagnaire

2008

Proceedings of the 2008 ACM Symposium on Applied Computing

View full text Add to dashboard Cite

Grid networks are large distributed systems that share and virtualize heterogeneous resources. Quality of Service (QoS) is a key and complex issue for Grid services provisioning. Currently, most Grid networks offer best-effort (BE) services. Thus, QoS architectures initially developed for Internet such as DiffServ (DS) have been adapted to Grid environment. Since the widespread of Internet, many Grid networks will be deployed in the years to come over this technology. In this paper, we propose to compare two Flow-Aware Networking (FAN) architectures, mainly from the second generation (2GFAN). The purpose is to answer the question of which 2GFAN architecture performs better under Grid traffic. FAN is a promising option to DS for QoS provisioning in Internet networks. DS provides QoS differentiation through explicit packet marking and classification whereas FAN consist on per-flow admission control and implicit flow differentiation through priority fair queuing. The main difference between the two 2GFAN architectures is the fair queuing algorithm. Thus; to the knowledge of the authors, this is the first time two priority per-flow fair queuing algorithms are compared under Grid traffic. A GridFTP session may be seen as a succession of parallel TCP flows with large volumes of data transfers. Metrics used are average delay, average goodput and the average rejection rate.

show abstract

Online Algorithms for Single Machine Schedulers to Support Advance Reservations from Grid Jobs

Li¹,

Zhao²

2007

High Performance Computing and Communications

View full text Add to dashboard Cite

QoS control method to reduce resource reservation failure in datagrid applications

Cited by 7 publications

References 1 publication

A large-scale data transfer method that reduces reservation bandwidth over DiffServ AF networks

A large-scale data transfer method that reduces reservation bandwidth over DiffServ AF networks

Performance comparison of Flow Aware Networking (FAN) architectures under GridFTP traffic

Online Algorithms for Single Machine Schedulers to Support Advance Reservations from Grid Jobs

Contact Info

Product

Resources

About