A Generic Cluster-Based Solution for Data Intensive Services

نویسنده

  • Amit Khanna
چکیده

A cluster-based architecture with a proxy and a few back-end servers is a popular solution to face the ever increasing demand for data-intensive services. Stateof-the-art Cluster-based servers employ a proxy that is responsible for distributing incoming requests to the back-end server in a manner that is transparent to clients. The back-end servers send the response to the proxy, which then forwards it to the client. The overhead of processing each packet at the proxy affects the response time adversely. Moreover, since the entire traffic from all the servers has to pass through the proxy, it often becomes the bottleneck. In this paper, we propose an architecture for cluster servers which provides a generic solution for the above mentioned problems. It is achieved by sending the response from the back-end server directly to the client, bypassing the proxy altogether. Thus, we eliminate the overhead of traversing the network stack at proxy which involves costly kernel-user context switches and data copies between the kernel and user space buffers. The technique used to achieve this is called Connection Handoff. The paper also argues the aptness of SCTP (Stream Control Transmission Protocol) as the transport layer protocol for our clusters. SCTP is similar to TCP but has additional features like multistreaming, multihoming, immunity to DoS attacks, prevention of Head of Line Blocking etc. These features make it a better choice as the transport layer protocol for our project. Using the technique of SCTP handoff along with features like Load balancers and Content-Aware-Dispatcher, we aim at improving user response times and overall throughput rates for a variety of data-intensive services, improving the scalability and robustness of the cluster servers at the same time.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Generic Approach to Satisfy Adaptability Needs in Mobile Environments

The need for redesigning existing “data-intensive” applications to execute in mobile environments has led to an enormous proliferation of systems that propose a solution to different issues. In this paper, we classify these approaches and give evidence of the lack for code reuse efforts. We introduce the MolèNE system as a set of generic services that aims at providing a systematic approach to ...

متن کامل

Cluster Based Cross Layer Intelligent Service Discovery for Mobile Ad-Hoc Networks

The ability to discover services in Mobile Ad hoc Network (MANET) is a major prerequisite. Cluster basedcross layer intelligent service discovery for MANET (CBISD) is cluster based architecture, caching ofsemantic details of services and intelligent forwarding using network layer mechanisms. The cluster basedarchitecture using semantic knowledge provides scalability and accuracy. Also, the mini...

متن کامل

Towards Benchmarking IaaS and PaaS Clouds for Graph Analytics

Cloud computing is a new paradigm for using ICT services— only when needed and for as long as needed, and paying only for service actually consumed. Benchmarking the increasingly many cloud services is crucial for market growth and perceived fairness, and for service design and tuning. In this work, we propose a generic architecture for benchmarking cloud services. Motivated by recent demand fo...

متن کامل

Direct Marketing Based on Fuzzy Clustering of Customers (Case Study: on one Mobile Company)

Objective There is a general tendency toward direct marketing these days. Therefore, instead of designing advertisement and marketing strategies for all the customers in the market, it is recommended to classify the customers based on clustering techniques and then design specific strategies accordingly. This will reduce marketing and advertisement expenses, increase sale department efficientl...

متن کامل

A Multi-core Fpga-based 2d-clustering Algorithm for High-throughput Data Intensive Applications

A multi-core FPGA-based clustering algorithm for high-throughput data intensive applications is presented. The algorithm is optimized for data with two dimensional organization (e.g. image processing, pixel detectors for high energy physics experiments etc.). It uses a moving window of generic size to adjust to the application’s processing requirements (the cluster sizes and shapes that appear ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2005