Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Priority based on-boarding between storage clouds

IP.com Disclosure Number: IPCOM000236893D
Publication Date: 2014-May-21
Document File: 4 page(s) / 37K

Publishing Venue

The IP.com Prior Art Database

Abstract

On-boarding is a cloud enablement technology, because it allows to new cloud providers to enter to the market and to attract the clients of the competitors to move to a new cloud. This disclosure describes the way to improve the user experience during the on-boarding process and to make the on-boarding to be more effective. This improvement is achieved by combining big-data analytics applied on a social data e.g. Facebook and web search data, with cloud meta-data and access paters and predicting user future behavior by observed patterns.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 30% of the total text.

Page 01 of 4

Priority based on -

On-boarding is a process in which a cloud customer migrates his
data from one storage cloud (an Old Cloud) to another cloud (a
New Cloud) [1]. The provider of the New Cloud is responsible for
moving the customer data to itself from the old cloud.
Furthermore, following the federation set up, applications and
users begin accessing their data through the new cloud, which
provides instant access to customer data remaining in the old
cloud. Thus, the applications that access the migrated data are
not influenced by on-boarding and can work transparently. The new
provider migrates the resources using a component called
Federator, gets the object from the old cloud and puts it in the
new cloud. From the technical standpoint, the on-boarding
architecture specifies the following three primary flows:

· On-boarding set-up: The process of setting the relationship
between a resource (e.g. a container) in in a new cloud and
a resource in an old cloud.

· Direct access: Once the relationship is set-up, all client's
applications may start to immediately access the objects of
the old container through the new cloud. When a client
accesses a resource in the new cloud that has not yet been
on-boarded, the Federator module gets the missing resource
from the old cloud and puts it in the new cloud.

· Background on-boarding: The Federator component creates
performs background on-boarding by fetch resources from the
old cloud and puting them to the new cloud. These jobs may
have lower priority and run when the resource utilization
(e.g., CPU and network) in the new cloud is low so that
they do not interfere with the normal operation of the
cloud.

Due to large amount of data that would be stored at Old Cloud and
network constraints it will take time to move all the data to new
cloud.

By prioritization of the data migration we can decrease the
response time for user data access.In a best case the response
time of accessing of the data that is still stored at Old Cloud
would be the sum of the response time of the old cloud and the
new one, but at some scenarios the New Cloud needs to get the
whole data object before it starts to response to the client GET
request, at such a cases the response time would be increased
dramatically.

    In addition if the new cloud has better site distribution
and its data centers are located near its clients then access of
the data that is already migrated would be faster, so the order
of data migration will effect the user experience.

    If a new cloud provides some analytic tools, then that
tools would access the data during there execution. And then if
not all the accessed data is already migrated to the New Cloud

-boarding between storage clouds

boarding between storage clouds


Page 02 of 4

such analytics would take a long.

    The conclusion is that prioritization of data migration
during the on-boarding process is very important, it would
improve the user experience, to reduce the response time for
internal processes and to meet the users' SLA.

    In d...