EFFICIENT TRUSTED CLOUD STORAGE USING PARALLEL CLOUD COMPUTING
1 like181 views
The document presents a prototype model for efficient and trusted cloud storage using parallel cloud computing, aimed at small and medium-sized businesses to optimize resource utilization at reduced costs. It addresses the limitations of direct attached storage (DAS) by introducing a distributed, shared storage solution that ensures high performance, fault tolerance, and eliminates single points of failure. The implementation of this model achieves 100% uptime and offers a viable platform for web hosting and cloud services.
![International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume 3, Issue 3 (May-June 2015), PP. 298-299
298 | P a g e
EFFICIENT TRUSTED CLOUD STORAGE USING
PARALLEL CLOUD COMPUTING
Shilpa Ozarkar1, Neetika Singh2
1
Student, 2
HOD (CSE Department)
DPGITM, Gurgaon
1
shilpa.ozarkar@gmail.com, 2
nitikasagi@gmail.com
Abstract- In last few years, rapidly increasing businesses
and their capabilities & capacities in terms of computing has
grown in very large scale. To manage business requirements
High performance computing with very large scale resources is
required. Businesses do not want to invest & concentrate on
managing these computing issues rather than their core business.
Thus, they move to service providers. Service providers such as
data centers serve their clients by sharing resources for
computing, storage etc. and maintaining all those.
To make optimal or maximum utilization of available resources,
service providers (data centers) use one of emerging technology
called, “Cloud Computing”. [1]
I. PURPOSE
This thesis work presents a prototype / model to small and
medium scale businesses, or hosting businesses, which allows
them to make maximum/optimal use of available resources
with reduced cost. This model will improve cloud server
availability and performance with elastic, scalable cloud
storage. Parallels Cloud Storage uses the existing unused disk
space on server nodes to deliver cloud storage at a fraction of
the cost of competing solutions. This model will provide no
single point of failure i.e. fault tolerant, high performance and
low cost.
II. SCOPE
Most service providers today provision their servers with
direct attached storage (DAS). While DAS is fast and
inexpensive to procure, it is has a high overall cost of
ownership and high failure rates that can significantly impact
service availability. In order to address the issues with DAS,
many service providers have considered storage area networks
(SAN) as an alternative; however these are typically too costly
for low cost hosting. [2]
In this thesis work, a prototype of Parallels Cloud Storage
reduces total operation cost and compares the capabilities and
relative cost of Parallels Cloud Storage to DAS and Storage
Area Network (SAN) alternatives.
Parallels Cloud Storage solves the problems associated
with DAS by providing a distributed, shared solution that
decouples storage from computation. This approach enables
virtual machines and containers to be instantly migrated to an
operational physical server if the original physical server
becomes unavailable.
III. INTRODUCTION
A. Objective
The main objective of this thesis work is to build and
present a prototype of Cloud Computing environment which
must fulfill high performance and high data availability of
resources as well as fast and flexible storage solution for
modern datacenters.
B. Architecture
Parallels Cloud Storage Architecture:
Fig 1.2 Typical parallel cloud storage Architecture
The basic component of Parallels Cloud Storage is a
cluster. The cluster is a group of physical computers
connected to the same Ethernet network and performing the
following roles:
C. Chunk Servers:
Chunk servers store all the data, including the contents of
virtual machines and Containers, in the form of fixed-size
chunks and provide access to these chunks.
D. Metadata Servers:
Metadata (MDS) servers store metadata about chunk
servers and control how files keeping the contents of virtual
machines and Containers are split into chunks and where these
chunks are located.
E. Clients:
Clients are computers with Parallels Cloud Server 6.0
from where you run virtual machines and Containers stored in
a Parallels Cloud Storage cluster. [3]
F. Parallel Cloud Computing - Brief Overview:
Parallel Cloud Server
Parallels Cloud Server is a virtualization solution that allows
you to simultaneously run Parallels virtual machines and](https://image.slidesharecdn.com/ijtra1505128-160123085357/85/EFFICIENT-TRUSTED-CLOUD-STORAGE-USING-PARALLEL-CLOUD-COMPUTING-1-320.jpg)
![International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume 3, Issue 3 (May-June 2015), PP. 298-299
299 | P a g e
containers on the same physical server, thereby making more
efficient use of your server's hardware resources. Parallel
Cloud Server is the only solution available today, that allows
deploying both containers and hypervisor on the same
physical server. This flexibility offers advantage of the
scalability and streamlined operations of operating system
virtualization (containers), while also making use of high-
performance virtual machines (hypervisors) for isolated
applications or simultaneous heterogeneous environments.
Parallel Cloud server has following features:
- Parallel Cloud Storage
- Container-in-an-image-file layout
- Memory and IOPS deduplication
- Rebootless Updates
- Console Access to containers
- Updates with yum
- Container with preinstalled application templates
- Virtual Machines with 32 CPU Cores, 128 GB of RAM, and
5 TB of Disk Space
- Support for Open vSwitches
- CPU identification masks for virtual machines
- Virtual Machine configuration samples. [4]
IMPLEMENTATION
Setting up of parallel cloud storage cluster includes
following steps:
A. Install Parallel Cloud Server 6.0 on Server [Master and
Slave]
B. Configuring Parallels Cloud Storage cluster discovery.
C. Checking data flushing.
D. Setting up Metadata Servers.
E. Setting up Chunk Servers.
F. Setting Up Clients.
G. Managing Parallel Cloud Storage Cluster using PVA
agent
IV. CONCLUSION
This thesis work helped to understand one of emerging
technology called, Cloud Computing and its pros and cons.
This project work helped to understand about cloud computing
in broad aspect including parallel computing and storage
technology. and areas which can be improve in the same
domain to meet newly and faster growing requirements of
businesses and solution which can be served, for long-time.
Being an ideal storage solution, Parallel Cloud Storage
has overcome single-point failures with much more efficient
use of existing storage. The ability of replication of data
chunks in parallel, across all servers in the cluster makes it
faster than RAID; also its reliability is much more.
The 3 MDS servers (1 Master and 2 Slave) & 3 chunk
servers shows automatic rebalancing of storage and recovery
of data from different chunks if any server goes down or fails.
Hence 100% uptime achieved with no single point failure
model. It is one of the low cost solution.
Implemented model also offers platform as a service with
many flavors of Linux based OS. Implemented server can also
be used for web hosting by assigning static IP address and
global DNS and registering it into Global DNS server.
Current model can be accessed in Local Area Network
only.
REFERENCES
[1] NIST Definition of Cloud Computing: NIST Special
Publication 800-145, September 2011. Page No. 6
http://csrc.nist.gov/publications/nistpubs/800-145/SP800-
145.pdf
[2] Whitepaper-on- Parallel Cloud Storage- The ideal Storage
Solution for Hosters, Page No. 3 [November2012].
http://www.parallels.com/fileadmin/parallels/documents/hosti
ng-cloud-
enablement/pcs/Production_Whitepapers/PCS_The_Ideal_St
orage_Solution_for_Hosters_WP_EN_Ltr_11282012.pdf
[3] Parallel Cloud Server - Key Features and Benefits [November
2012]
http://www.parallels.com/fileadmin/parallels/documents/hosti
ng-cloud-
enablement/pcs/Production_Whitepapers/PCS_Key_Features
_and_Benefits_WP_Ltr_EN_113012.pdf
[4] New Features and improvements:
http://www.parallels.com/products/pcs/](https://image.slidesharecdn.com/ijtra1505128-160123085357/85/EFFICIENT-TRUSTED-CLOUD-STORAGE-USING-PARALLEL-CLOUD-COMPUTING-2-320.jpg)
EFFICIENT TRUSTED CLOUD STORAGE USING PARALLEL CLOUD COMPUTING
- 1. International Journal of Technical Research and Applications e-ISSN: 2320-8163, www.ijtra.com Volume 3, Issue 3 (May-June 2015), PP. 298-299 298 | P a g e EFFICIENT TRUSTED CLOUD STORAGE USING PARALLEL CLOUD COMPUTING Shilpa Ozarkar1, Neetika Singh2 1 Student, 2 HOD (CSE Department) DPGITM, Gurgaon 1 shilpa.ozarkar@gmail.com, 2 nitikasagi@gmail.com Abstract- In last few years, rapidly increasing businesses and their capabilities & capacities in terms of computing has grown in very large scale. To manage business requirements High performance computing with very large scale resources is required. Businesses do not want to invest & concentrate on managing these computing issues rather than their core business. Thus, they move to service providers. Service providers such as data centers serve their clients by sharing resources for computing, storage etc. and maintaining all those. To make optimal or maximum utilization of available resources, service providers (data centers) use one of emerging technology called, “Cloud Computing”. [1] I. PURPOSE This thesis work presents a prototype / model to small and medium scale businesses, or hosting businesses, which allows them to make maximum/optimal use of available resources with reduced cost. This model will improve cloud server availability and performance with elastic, scalable cloud storage. Parallels Cloud Storage uses the existing unused disk space on server nodes to deliver cloud storage at a fraction of the cost of competing solutions. This model will provide no single point of failure i.e. fault tolerant, high performance and low cost. II. SCOPE Most service providers today provision their servers with direct attached storage (DAS). While DAS is fast and inexpensive to procure, it is has a high overall cost of ownership and high failure rates that can significantly impact service availability. In order to address the issues with DAS, many service providers have considered storage area networks (SAN) as an alternative; however these are typically too costly for low cost hosting. [2] In this thesis work, a prototype of Parallels Cloud Storage reduces total operation cost and compares the capabilities and relative cost of Parallels Cloud Storage to DAS and Storage Area Network (SAN) alternatives. Parallels Cloud Storage solves the problems associated with DAS by providing a distributed, shared solution that decouples storage from computation. This approach enables virtual machines and containers to be instantly migrated to an operational physical server if the original physical server becomes unavailable. III. INTRODUCTION A. Objective The main objective of this thesis work is to build and present a prototype of Cloud Computing environment which must fulfill high performance and high data availability of resources as well as fast and flexible storage solution for modern datacenters. B. Architecture Parallels Cloud Storage Architecture: Fig 1.2 Typical parallel cloud storage Architecture The basic component of Parallels Cloud Storage is a cluster. The cluster is a group of physical computers connected to the same Ethernet network and performing the following roles: C. Chunk Servers: Chunk servers store all the data, including the contents of virtual machines and Containers, in the form of fixed-size chunks and provide access to these chunks. D. Metadata Servers: Metadata (MDS) servers store metadata about chunk servers and control how files keeping the contents of virtual machines and Containers are split into chunks and where these chunks are located. E. Clients: Clients are computers with Parallels Cloud Server 6.0 from where you run virtual machines and Containers stored in a Parallels Cloud Storage cluster. [3] F. Parallel Cloud Computing - Brief Overview: Parallel Cloud Server Parallels Cloud Server is a virtualization solution that allows you to simultaneously run Parallels virtual machines and
- 2. International Journal of Technical Research and Applications e-ISSN: 2320-8163, www.ijtra.com Volume 3, Issue 3 (May-June 2015), PP. 298-299 299 | P a g e containers on the same physical server, thereby making more efficient use of your server's hardware resources. Parallel Cloud Server is the only solution available today, that allows deploying both containers and hypervisor on the same physical server. This flexibility offers advantage of the scalability and streamlined operations of operating system virtualization (containers), while also making use of high- performance virtual machines (hypervisors) for isolated applications or simultaneous heterogeneous environments. Parallel Cloud server has following features: - Parallel Cloud Storage - Container-in-an-image-file layout - Memory and IOPS deduplication - Rebootless Updates - Console Access to containers - Updates with yum - Container with preinstalled application templates - Virtual Machines with 32 CPU Cores, 128 GB of RAM, and 5 TB of Disk Space - Support for Open vSwitches - CPU identification masks for virtual machines - Virtual Machine configuration samples. [4] IMPLEMENTATION Setting up of parallel cloud storage cluster includes following steps: A. Install Parallel Cloud Server 6.0 on Server [Master and Slave] B. Configuring Parallels Cloud Storage cluster discovery. C. Checking data flushing. D. Setting up Metadata Servers. E. Setting up Chunk Servers. F. Setting Up Clients. G. Managing Parallel Cloud Storage Cluster using PVA agent IV. CONCLUSION This thesis work helped to understand one of emerging technology called, Cloud Computing and its pros and cons. This project work helped to understand about cloud computing in broad aspect including parallel computing and storage technology. and areas which can be improve in the same domain to meet newly and faster growing requirements of businesses and solution which can be served, for long-time. Being an ideal storage solution, Parallel Cloud Storage has overcome single-point failures with much more efficient use of existing storage. The ability of replication of data chunks in parallel, across all servers in the cluster makes it faster than RAID; also its reliability is much more. The 3 MDS servers (1 Master and 2 Slave) & 3 chunk servers shows automatic rebalancing of storage and recovery of data from different chunks if any server goes down or fails. Hence 100% uptime achieved with no single point failure model. It is one of the low cost solution. Implemented model also offers platform as a service with many flavors of Linux based OS. Implemented server can also be used for web hosting by assigning static IP address and global DNS and registering it into Global DNS server. Current model can be accessed in Local Area Network only. REFERENCES [1] NIST Definition of Cloud Computing: NIST Special Publication 800-145, September 2011. Page No. 6 http://csrc.nist.gov/publications/nistpubs/800-145/SP800- 145.pdf [2] Whitepaper-on- Parallel Cloud Storage- The ideal Storage Solution for Hosters, Page No. 3 [November2012]. http://www.parallels.com/fileadmin/parallels/documents/hosti ng-cloud- enablement/pcs/Production_Whitepapers/PCS_The_Ideal_St orage_Solution_for_Hosters_WP_EN_Ltr_11282012.pdf [3] Parallel Cloud Server - Key Features and Benefits [November 2012] http://www.parallels.com/fileadmin/parallels/documents/hosti ng-cloud- enablement/pcs/Production_Whitepapers/PCS_Key_Features _and_Benefits_WP_Ltr_EN_113012.pdf [4] New Features and improvements: http://www.parallels.com/products/pcs/