Posts: 5,362
Threads: 2,998
Joined: Feb 2011
Prepared by
T.Rama Rao
[attachment=10372]
Quiver:Consistent Object Sharing for Edge Services
The principle of a Web service.
• A Web service is a software system designed to support interoperable machine-to-machine interaction over a network.
• Interoperability is a property referring to the ability of diverse systems and organizations to work together (inter-operate), impact system to system performance The IEEE defines interoperability as: the ability of two or more systems or components to exchange information and to use the information that has been exchanged.James A. O'Bien and George M. Marakas define interoperability as: Being able to accomplish end-user applications using different types of computer systems, operating systems, and application software, interconnected by different types of local and wide area networks.
• Web services today are frequently just Application programming interface (API) or web APIs that can be accessed over a network, such as the Internet, and executed on a remote system hosting the requested services. Extensible Markup Language (XML) or JavaScript Object Notation (JSON) format.
Web services in a service-oriented architecture.
• More recently, REpresentational State Transfer (RESTful) web services have been regaining popularity, particularly with Internet companies. By using the PUT, GET and DELETE http methods, alongside POST, these are often better integrated with http and web browsers than SOAP-based services. They do not require XML messages.
• Web API is a development in web services (in a movement called Web 2.0) where emphasis has been moving away from Simple Object Access Protocol (SOAP) based services towards more direct Representational State Transfer (REST) style communications.
Styles of use:
Web service is a set of tools that can be used in a number of ways. The
3most common styles of use are RPC, SOA, and REST.
1.RPC web service present a distributed function/method call interface that is familiar to many developers.
2.SOA(service oriented architecture): web service can also be used to implement an architecture according to soa.
3.Representational State Transfer(rest) attempts to describe architectures which use http or similar protocols by constraining the interface to a set of well known,standard operations like get,post,put,delete for http.
• Web services provide access to software systems over the Internet using standard protocols. In a minimalistic scenario there exists at least a Web service provider that publishes some service such as a weather service and a Web service consumer that uses this service. Web service discovery is the process of finding a suitable Web service for given task.
• Publishing a Web service involves at the bare minimum to create the software artifact and make it accessible to potential consumers. In order that a consumer can use a service, providers usually augment a Web service endpoint with an interface description using the Web Services Description Language (WSDL).
• A provider can explicitly register web service in web services registry such as Universal Description Discovery & Integration.
• WEB SERVICE DESIGN METHODOLOGIES
Web service can be written in 2 ways.
1.Using the “bottom up method”A developer first writes the implementation class in a programming language & then uses wsdl generating tool to expose its methods as a web service.This is often the simpler approach.
2.Using the “top down method” ,a developer first writes wsdl document and then uses a code generating tool to produce the class skelton.
• Existing Network
The principle of cooperative caching.
Regarding web proxy or proxy Server
• A web proxy sits between corporate client machines and the Internet. Sometimes it is a stand-alone product; more often, it is functionality built into other software. A forward proxy stands in the way of the Internet and receives data from client machines bound for the Internet and forwards it on on their behalf. A reverse proxy stands in the way of servers within the corporate network and receives data bound for them, and distributes it on.
• Proxy servers sit between a client program (typically a Web browser) and an external server (typically another server on the Web) to filter requests, improve performance, and share connections.
• In computer networks, a proxy server is a server (a computer system or an application program) that acts as an intermediary for requests from clients seeking resources from other servers. A client connects to the proxy server, requesting some service, such as a file, connection, web page, or other resource, available from a different server. The proxy server evaluates the request according to its filtering rules. For example, it may filter traffic by IP address or protocol. If the request is validated by the filter, the proxy provides the resource by connecting to the relevant server and requesting the service on behalf of the client. A proxy server may optionally alter the client's request or the server's response, and sometimes it may serve the request without contacting the specified server. In this case, it 'caches' responses from the remote server, and returns subsequent requests for the same content directly.
This system employs service proxies at the edge of internet to serve
distributed clients accessing a service involving mutable objects.In this System,
service proxies(web proxies) are connected with server in critical structure. Here
server provides services to proxies.when user sends request for an object,if that
object is available at any proxy,that proxy only serves to that request.So user
request need not go to server which may be busy with other services.
CAPABILITIES/MERITS:-
1) distributing the operation processing load among the service-proxies.
2) enabling clients to access the service by communicating with nearby proxies rather than a potentially centralized server.
Here service proxies perform sequential access to shared objects.These
shared objects are migrated to proxies as migration is an essential thing in distribu
ted computing/system.Then proxies perform operations on those shared objects.
Service proxy handles requests from clients on behalf of server. Service proxies get
services from server.Proxies only serves to clients’requests.Service enables proxies
to invoke operations on objects.These operations may be reads or updates.Updates
compute object instances from other object instances.
• This approach has the potential of both distributing the operation processing load among the proxies and enabling clients to access the service by communicating with nearby proxies rather than a potentially distant centralized server.
• A major challenge in this architecture, however, is to enable the (globally distributed) service proxies to efficiently access the mutable service objects for servicing client operations while ensuring strong consistency semantics for these object accesses.
• Applications
• Stock Prices.
• Weather Reports.
PROBLEM WITH PRESENT SYSTEM/DE MERITS
Object migration to dynamic services does not handle failure of proxies/proxy disconnections With centralized server,Each proxy should be connected with.The consistency is not maintained while sharing object between proxies because proxies can alter objects.Object is lost if proxy fails.
In the existing system the proxies have been maintained in the sequence of network
(critical path).
For each object updation or each proxy should be connected with the
centralized server.
The consistency is not maintained while sharing the object.It will make the sharing
process inconsistent.
It doesnot handle proxy disconnections.If the proxy has failed means the object has
been lost.
The existing system supports only single-object operations.
Where as in proposed system,proxies have been maintained in tree structure using
kruskal algorithm.Kruskal algrtm generates a minimum-cost spanning tree for every
connected undirected proxies tree/graph.Algrthm sorts costs in ascending ordert and
generates minimum cost spanning tree.Kruskal’s algrtm reduces weightage in tree
structure and workload in the server.
Consistency means ability to repeat a particular task with minimal variation.Consistent
object sharing among the proxies enables them to export the same consistent view of
the service to the clients in turn. However, achieving even just serializability for
operations executed at these proxies using standard replication approaches requires
that a proxy involve either a centralized server or other (possibly distant) proxies on the
critical path of each update operation. Strict serializability techniques require wide-area
interactions for reads as well.
Where as Quiver system provides an alternative approach to achieving consistent access
to objects by edge proxies while retaining the proxies load dispersing and latency redu
cing effects.Quiver organizes the proxies in a tree rooted at the server.The tree is
structured so that geographically close proxies reside close to one another in the tree.To
perform certain types of operations,a proxy uses the tree to migrate each involved object
to itself and then performs the operation locally.Existing system::drawback of object
migration is increased sensitivity to proxy disconnections::if a proxy disconnects while
holding an object because the proxy fails,it can no longer communicate with its parent,
or its parent disconnects,then operations that it recently applied to object may be lost.
Wher as in Quiver system,the connected component of the tree containing the server
can regenerate the last version of the object seen in that component when such a
disconnection is detected.Thus,server never loses control of the service objects, and
once an object reaches a portion of the tree that stays connected,all operations that it
reflects become durable.Alternatively,an operation can be forced to be durable when
performed but at additional cost.For these durable operations,quiver can implement
either serializability or strict serializability.
Purpose:
• Project major intention is to assign N number of substitutes (proxies) in the form of tree structure and to share the applications (objects) from server to proxies. Edge Services is an overlay infrastructure which extends appliances to operate on and provide a value-add to the content to be delivered and it will perform tasks instead of server. In the existing sharing process, only single object operations are retained and proxies have been maintained in the sequence of network (critical path). Proxy will be connected with the centralized server so if the proxy fails, object will be lost. It will make the sharing process inconsistent. And it is not capable of handling the proxy disconnections.
• To get rid of this problem we propose a system with a tree structure, which is maintained by kruskal algorithm in order to reduce the weightage. Tree structure will help to share the objects and to reduce the workload of the server. It will enable the system to do consistent multi object operations. If the proxy is disconnected, object will be holded till the alternate connection establishment. As a result highly scalable object sharing process is complete with strong consistency and multi object operations are performed while supporting more efficient single-object reads.
Scope:
• This project will be applicable in multi-tier architecture or in client server transaction. Our protocol supports decentralized proxy and flexible delegation of services.
• This system forms the service proxies in the tree structure. It shares the objects within the proxies. It reduces the workload in the server.
