OpenStack and Object Storage(Swift )

OpenStack Platform

Openstack Is an open source cloud computing platform used to implement the cloud environment. This environment is providing the best way to manage a large amount of data, it has a large number of resources, networking, and storage which provides reliable performance. The OpenStack platform will accept all integrations and the operating system going to be used is Cloud Admin’s choice. In OpenStack, there are few services required to build the cloud. The Services are,Keystone which is used for authentication purpose, Glance is an Image service which is providing the image service to the Instance,Nova is a compute service which run the instance ,Neutron is a networking service which manages the networks,and the Horizon is a Dashboard service which provides the user interface for managing the Cloud. The cloud resources are managed by the users based on their requirements through the dashboard or OpenStack CLI or API.The enterprise data files could be created, uploaded and downloaded in the cloud storage through the Dashboard to meet the needs of remote storage, backup, archiving and information share for the enterprise users and personals.

Object Storage

Object storage combines the features of the block storage and file storage, so the data reading will be fast. In object storage, every data is stored on an object that contains the metadata and the unique identifier. There is no limit for metadata, which is making object storage powerful. In Distributed systems the object storage stores the multiple copies of the data, so the objects are protected.

In object storage we can store unlimited data as an object, each object has its own URL for identification and we have a rights to configure the maximum size of the object. In object storage, we can compress files using content-metadata and auto extraction is available on the object storage. We have the rights to set our own storage policies and replication settings in the object storage. With the help of the metadata, we can manage the files with the high scalability. We can access and manage large amountOperating system is not required to access the object storage, the object storage accessed through Restful APIs using HTTP(S) web Protocol. The cloud resources are managed by the users based on their requirements through the dashboard.

Object storage provides a solution to manage a large amount of unstructured data. It has the advantages of high data persistence, completely symmetrical system architecture, and unlimited scalability, no single point of failure, simple and based on the reliable component. of data at low cost using object storage and access this through a Restful API


Swift is a OpenStack subproject used to build an object storage environment. It is implemented through a few separate services which ddeliverthe capacities of the object storage.

Object Upload in Swift(Object Storage)


It will responsible for processing the API requests and other components communicate with each other. It also accepts the API and the other HTTP requests for creating and maintaining containers, upload and maintaining the files in the object storage, shows the container listings in the web browsers. The proxy-server was used the optional cache which was deployed with the Memcache for improving the performance. It’s a physical form to be a single server or extended to a proxy-server cluster for improving the load balance capacity of a system.


Account-server will manges the container listings,the Container Server will manages the object listings and it will not know the information about where the objects are,but knows the what objects in the specific container. The Object-server used to store, retrieve and delete objects.

Swift Features

Using object storage is a suitable solution to manage the rapid growth of the digital data,by using the object storage we can manage the large amount of enterprise’s data at low cost by Restful API. The main advantage of the object storage metadata feature, it was stored with the object data itself, and was used as a unique id to retrieve the objects easily. For example if we store a audio file with its name,author name,released year and other details about that file,these details are stored along with its object file and there is no need of external database to store that details as a metadata.


In Object Storage each replica is working Independently, and the replicas diverged by transient failures which includes the network Partition.These Differences are compromised by the asynchronous p2p replicators.Push model is used by replications,the files and records are being copied from local to remote replicas,the replicators are does not know about the data which was pull from anywhere in the cluster.The replica was handled by the ring.Two types of the replications is available in object storage.

1.Database Replication

2.Object Replication    

1.Database Replication

In this replication Container and object is replicated

2.Object Replication

In this replication object data is replicated