Note: Descriptions are shown in the official language in which they were submitted.
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
INTEGRATED COMPUTING PLATFORM DEPLOYED IN AN EXISTING COMPUTING
ENVIRONMENT
BACKGROUND OF THE INVENTION
Field of the Invention
[0001] Embodiments of the present invention generally relate to a method
for
installing a computing block platform within an existing data center.
Description of the Related Art
[0002] Generally, in a virtualized computer environment, virtual
machines are
configured to run on one or more host computers. Each virtual machine uses CPU
and
memory resources of one of the host computers and runs its own operating
system and
application programs to give the appearance that it is operating as a stand-
alone
computer system. The amount of CPU and memory resources provisioned for each
of
the virtual machines on a host computer can be designated by an administrator
of the
virtualized computer environment. In some virtualized environments, load
balancing is
enforced across multiple host computers by software that monitors resource
usage on
different host computers. Such software migrates virtual machines from one
host
computer to another, e.g., from a busy host computer to one that has excess
capacity.
[0003] Additionally, enterprises engaged in developing, testing, and
deploying
software applications need to deal with many layers of the targeted platform.
These
layers include application services, virtualization, and hardware
infrastructure with
compute, network, storage, and management at all levels. Information
technology (IT)
and engineering groups also acquire, deploy, and provide ongoing management,
as well
as ensure the layers work seamlessly together. This increases an enterprise's
initial
and ongoing cost, extends the development cycle, and reduces flexibility
needed to
respond to changes in the market.
[0004] Further, the traditional enterprise information technology (IT)
roles such as
server administrator, UNIX administrator, and network, storage or exchange
administrator, have been generally static. The roles operate in isolation, or
"silos",
which cause friction within IT organizations, as well as between developers
and IT.
Further, it is difficult for developers to play the role of IT administrator
when their
developed applications are deployed as "cloud applications." This difficulty
is due not
1
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
only to various access-controls, authentication and authorization
complexities, but also
to a model of organizational separation between developers and IT, where the
default
behavior is to deny access to the resources.
[0005] As such, there is a demand for a more efficient operational model
for
administrating computing infrastructure.
SUMMARY OF THE INVENTION
[0006] Embodiments of the present disclosure provide a method for
installing an
integrated computing platform into an existing computing environment. The
method
includes configuring an integrated computing platform to communicate with at
least one
service running in an existing computing environment outside of the integrated
computing platform. The integrated computing platform includes an application
having a
plurality of functional components using a plurality of networking, storage,
computing,
and virtualized resources of the integrated computing platform. The method
further
includes restoring each of the plurality of networking, storage, computing,
and virtualized
resources of the integrated computing platform to a state prior to the
configuring the
integrated computing platform to communicate with the service running in the
existing
computing environment outside of the integrated computing platform.
[0007] Embodiments of the present disclosure provide an integrated
computing
platform. The integrated computing platform includes a plurality of
networking, storage,
computing, and virtualized resources having an application with a plurality of
functional
components executing thereon. The integrated computing platform includes a
management server configured to configure the integrated computing platform to
communicate with at least one service running in an existing computing
environment
outside of the integrated computing platform. The management server is further
configured to restore each of the plurality of networking, storage, computing,
and
virtualized resources of the integrated computing platform to a state prior to
the
configuring the integrated computing platform to communicate with the service
running
in the existing computing environment outside of the integrated computing
platform.
[0008] Embodiments of the present disclosure provide a non-transitory
computer-
readable storage medium comprising instructions that, when executed in a
computing
2
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
device, install an integrated computing platform into an existing computing
environment,
by performing the steps of configuring an integrated computing platform to
communicate
with at least one service running in an existing computing environment outside
of the
integrated computing platform. The integrated computing platform may include
an
application having a plurality of functional components using a plurality of
networking,
storage, computing, and virtualized resources of the integrated computing
platform. The
instructions, when executed, also perform the steps of restoring each of the
plurality of
networking, storage, computing, and virtualized resources of the integrated
computing
platform to a state prior to the configuring the integrated computing platform
to
communicate with the service running in the existing computing environment
outside of
the integrated computing platform.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] So that the manner in which the above recited features of the
present
invention can be understood in detail, a more particular description of the
invention,
briefly summarized above, may be had by reference to embodiments, some of
which are
illustrated in the appended drawings. It is to be noted, however, that the
appended
drawings illustrate only typical embodiments of this invention and are
therefore not to be
considered limiting of its scope, for the invention may admit to other equally
effective
embodiments.
[0010] Figure 1 illustrates an integrated computing platform configured for
providing
a virtualized environment according to one embodiment.
[0011] Figure 2 illustrates an application-specific deployment of the
integrated
computing platform of Figure 1, according to one embodiment.
[0012] Figure 3 is a flow diagram of method steps for deploying an
integrated
computing platform in an existing data center, according to one embodiment.
[0013] Figure 4 is a flow diagram of method steps for capturing state of
an integrated
computing platform, according to one embodiment.
[0014] Figure 5 is a flow diagram of method steps for restoring
configuration changes
to an integrated computing platform installed in an existing data center,
according to one
embodiment.
3
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0015] Figure 6 is a state diagram illustrating a workflow for resetting
state of an
integrated computing platform during installation of an application, according
to one
embodiment.
[0016] Figure 7 illustrates an example operation for deploying an
application in the
computing platform of Figure 1, according to one embodiment.
[0017] Figure 8 is a flow diagram of a method for deploying an
application on
computer resources in a virtualized environment, according to one embodiment.
[0018] Figure 9 is a flow diagram of a method for integrating a
computing platform
having an application executing thereon into an existing computing
environment,
according to one embodiment.
[0019] Figure 10 illustrates an example workflow for integrating the
computing
platform of Figure 2 within an existing data center, according to one
embodiment.
[0020] To facilitate understanding, identical reference numerals have
been used,
where possible, to designate identical elements that are common to the
figures. It is
contemplated that elements disclosed in one embodiment may be beneficially
utilized on
other embodiments without specific recitation.
DETAILED DESCRIPTION
[0021] Embodiments of the present disclosure use a computing block
platform,
sometimes referred to as converged infrastructure, that consolidates IT
infrastructure
into a pool of virtualized physical as well as computing, storage, and
networking
capacity that is shared by multiple application and lines of businesses have
been
proposed to address the problem of silo architectures and IT sprawl. When an
enterprise uses a converged infrastructure platform, the enterprise may be
faced with
the challenge of integrating the new converged infrastructure platform within
any
existing computing infrastructure, such as a data center. Before a converged
infrastructure platform can be used by an enterprise, the physical resources
(e.g., blade
servers, network switches, storages) and virtual machines that are "inside"
the
converged infrastructure platform have to be configured to communicate with
physical
resources and virtual machines that are "outside" the configured
infrastructure platform,
4
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
and vice versa. Further, a converged infrastructure platform may be configured
to run
an application having multiple functional components (e.g., executing on
multiple virtual
machines). At least one of the functional components may rely on existing
services
running outside of the converged infrastructure, and must be configured to
communicate
with any required services running in the existing data center.
[0022] As such, to install a converged infrastructure platform having an
application
pre-loaded therein, a number of changes may have to be made to the networking,
storage, and computing resources of the converged infrastructure platform. For
example, the network within the converged infrastructure has to understand the
data
center network already in place, extend the network setup into the converged
infrastructure platform, and enable communication between the converged
infrastructure
and the rest of the network in the data center. To complete installation, the
converged
infrastructure platform may create new VLANs and port groups, as well as
change new
virtual machines and edit settings of existing virtual machines.
[0023] However, if the installation fails (e.g., unable to connect with the
network of
the existing data center), it may be desirable to "rollback" changes done to
the system
such that the installation may start again from a clean state. Accordingly,
embodiments
of the present disclosure provide an administrative application that generates
snapshots
of the various resources (e.g., networking, storage, computing, virtual) of
the converged
infrastructure platform before and after configuration for an application. The
administrative application may initiate a rollback operation that restores
configurations to
the networking, storage, computing, and virtual resources. A rollback or reset
of the
application means not only reverting to known snapshots on the virtual
infrastructure
(e.g., virtual machines), but also roll back of configurations changes of
physical
resources to free up those resources for future needs.
[0024] Additionally, manually deploying an application on any such
converged
infrastructure is error prone, as doing so typically requires working
knowledge of both
the converged infrastructure and the particular application to be deployed.
Accordingly,
embodiments presented herein include a packaged application delivery mechanism
to
deploy applications having multiple functional components, which themselves
may
require complex configuration and setup. For example, in one embodiment, a
deterministic approach is used to allocate resources per application within a
pool of
5
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
resources from the computing block platform. The application can be layered on
the
allocated resources and delivered in a prepackaged form. The packaged
application
delivery mechanism discovers a set of available converged infrastructure
resources and
determines the resources required for a given application that is to be
deployed within
the computing block platform. The mechanism provides a model that deploys
services,
functional components, tiers, nodes, etc. required by the application layer on
top of the
allocated services.
[0025] Further, a converged infrastructure platform may be pre-
configured to run an
application having multiple functional components. At least one of the
functional
components may rely on existing services running outside of the converged
infrastructure. As such, certain functional components running in the
converged
infrastructure platform must be configured to communicate with any required
services
running in the existing data center (e.g., outside of the converged
infrastructure
platform). Embodiments presented herein use an infrastructure template that
describes
the existing computing environment (e.g., the data center). The converged
infrastructure platform uses the information specified in the infrastructure
template to
configure its physical resources and virtual resources (e.g., VMs) to allow
communication between the converged infrastructure and the rest of the data
center.
Accordingly, embodiments presented herein reduce the time to integrate
components of
a converged infrastructure with services of a data center. Further,
embodiments
presented herein reduce the risk of error during configuration of network
resources and
security services,
[0026] Figure 1 illustrates an integrated computing platform 102
configured to
provide a virtualized environment, according to one embodiment of the present
disclosure. A system administrator 150 desires to deploy the integrated
computing
platform 102 within an existing computing environment (e.g., data center 100).
The data
center 100 may include a plurality of servers (illustrated as servers 1041,
1042, 104m)
that run one or more services 106. It should be recognized that the servers
104 may
include conventional computing components (e.g., processor, memory, storage)
or may
be virtual machines (VMs) executing on such physical hardware. The services
106
running on the servers 104 provide one or more IT functions within the data
center,
including directory services, web server, database server, accounting,
application
6
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
serving, file management, storage, backup services, etc. As described in
detail below,
the system administrator may wish to deploy the integrated computing platform
102
such that the physical resources and virtual resources (e.g., VMs) running
inside of the
integrated computing platform 102 may communicate with the services 106 of the
existing data center 100.
[0027] As shown, the computing platform 102 includes a physical
infrastructure 110
configured to support a virtualized infrastructure 120. In the embodiment
shown in
Figure 1, physical infrastructure 110 includes hardware resources, such as
servers 1161
to 116N (sometimes referred to as "hosts") and one or more storage array
networks
(SAN), such as SAN 118, connected by a network 114. The virtualized
infrastructure
120 may include a virtualization environment 124 which itself includes one or
more
virtual machines 140. The computing platform 102 may be connected to other
computing systems, such as workstations, personal computers, data center
servers, via
a network 128, such as the Internet. In one embodiment, components of the
computing
platform 102 (e.g., servers, network, storage, software, etc.) may be
organized into a
single integrated framework sometimes referred to as "converged
infrastructure.," The
components of the computing platform 102 provide a pool of virtualized server,
storage,
network resources shared by multiple applications and/or organizations within
an
enterprise.
[0028] According to one embodiment, the physical infrastructure 110 may be
organized into a "computing-block" based infrastructure, wherein physical
infrastructure
units are characterized by repeatable units of construction having similar
performance,
operational characteristics, and discrete requirements of power, space, and
cooling that
facilitate rapid deployment, integration, and scalability. For example, the
"computing
block" may dynamically provision hardware resources based on performance
demands
placed on the physical infrastructure 110. One example of physical
infrastructure 110 is
a VblockTM System available from the VCE Company, LLC.
[0029] The physical infrastructure 110 further includes an
infrastructure manager 112
configured to manage the configuration, provisioning, and policy compliance of
the
physical infrastructure 110. Infrastructure manager 112 provides an interface
to
manage the provisioning of hardware resources (e.g., computing, networking,
storage)
with policy-based automation. According to one embodiment, the infrastructure
7
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
manager 112 may be included in each unit of physical infrastructure 110 to
manage the
configuration, provisioning, and compliance of each distinct computing block.
The
infrastructure manager 112 may simplify deployment and integration into IT
service
catalogs and workflow engines, and dramatically simplifies computing-block
platform
deployment by abstracting the overall provisioning while offering granular
access to
individual components for troubleshooting and fault management.
[0030] In one embodiment, the infrastructure manager 112 may include a
configuration including a list of IP address and system credentials to assign
newly
provisioned systems. The platform manager 130 and/or the virtualized
infrastructure
120 may connect to and communicate with the infrastructure manager 112 of the
physical infrastructure 110 to manage and/or configure the physical
infrastructure 110.
One example of an infrastructure manager includes EMC lonix Unified
Infrastructure
Manager (UIM) available from EMC Corporation. Similarly, the network 114 may
include a network manager configured to configure network devices (e.g.,
switches,
routers) and manage addressing, subnets, virtual local area networks (VLANs),
and
other network configurations. One example of a network manager includes a
Cisco
Switch accessible via a Cisco IOS command line interface (CLI) available from
Cisco
System, Inc.
[0031] The virtualized infrastructure 120 includes a virtualization
environment 124
configured to simulate (i.e., to virtualize) conventional components of a
computing
device, e.g., a processor, system memory, a hard disk drive, for executing one
or more
virtual machines 140. For example, each virtual machine 140 may include a
virtual
processor and a virtual system memory configured to execute an application. In
one
example implementation of an embodiment similar to that of Figure 1,
virtualization
environment 124 may be implemented by running VMware vSphere - or VMware
ESXabased hypervisor technologies on servers 1161 to 116n provided by VMware,
Inc.
of Palo Alto, California (although it should be recognized that any other
virtualization
technologies, including Xen and Microsoft Hyper-V virtualization technologies
may be
utilized consistent with the teachings herein). As described above, a
hypervisor
application may provide a foundation for building and managing a virtualized
IT
infrastructure, such as the virtualized infrastructure 120. The hypervisor may
abstract
8
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
processor, memory, storage and networking resources into multiple virtual
machines
that run unmodified operating systems and applications.
[0032] In one embodiment, the virtualized infrastructure 120 may be
managed by a
virtualization manager 122 (e.g., implemented as a process running in a
virtual machine
in one embodiment). In one embodiment, the virtualization manager 122 may
provide
end-to-end datacenter management through a set of APIs that enable integration
with
third-party management tools. The virtualization manager 122 may be configured
to
manage provision of VMs 140 within the virtualized infrastructure 120 and to
configure
VMs 140 with computing, network, and storage configurations suitable for
interoperability with other VMs 140 within the virtualized infrastructure 120.
One
example of the virtualization manager 122 may be the VMware vCenter
virtualized
management platform from available from VMware, Inc.
[0033] As shown, the computing platform 102 also includes a platform
manager 130
connected to the virtualized infrastructure 120 and physical infrastructure
110 by the
communications network. The platform manager 130 is configured to provision
and
configure resources from physical infrastructure 110 and virtualized
infrastructure 120
for use in deploying an application within the computing platform 102. For
example, if
virtualized infrastructure 120 requires additional VMs to scale a currently
running
application during peak traffic, platform manager 130 can coordinate with the
virtualization manager 122 to instantiate additional virtual machines to
support such
needs and configure the instantiated virtual machines with network settings
matching
those existing virtual machines. In another example, the platform manager 130
may
modify an existing VM 140 to add, remove, or configure virtual resources, such
as a
virtual network interface card (vNIC), assigned to the VM.
[0034] As shown, the platform manager 130 includes a snapshot service 132
configured to save a configuration state of the computing platform 102, across
all levels
of resources of the computing platform 102, including networking resources
(e.g.,
network 114), storage resources (e.g., SAN 118), computing resources (e.g.,
servers
116), and virtual resources (e.g., VMs 140). In one embodiment, the snapshot
service
132 may save the configuration state as "snapshots" 138 of the networking,
storage,
computing, and virtual machines. The type of a snapshot 138 may vary depending
on
the type of resource captured. For example, a snapshot of a VM 140 may
comprise a
9
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
container file that encapsulates the state of a VM including its documents and
data, and
information about its virtual hardware (e.g., CPU, memory, disk, etc.); a
reference or link
to a container file; or a delta file describing differences between one or
more container
files. In another example, a snapshot of a networking resource (e.g., network
114)
includes a log of administrative operations performed on the networking
sources, such
as adding VLANs, creating DMZs, etc.
[0035] As shown, the snapshot service 132 may maintain a repository of
snapshots
138 that may be later used to restore configuration changes to the resources
of the
computing platform 102. In one embodiment, the repository of snapshots 138 may
be
implemented as a repository of operations performed on the networking,
storage,
computing, and virtual resources of the computing platform 102. In operation,
the
snapshot service 132 may add, remove, and update entries in the repository as
needed.
For example, the snapshot service 132 may purge a set of entries in the
repository of
snapshots 138 in response to determining a successful installation and
deployment of
the computing platform 102 within the data center 100, the set of entries now
extraneous.
[0036] In one embodiment, the platform manager 130 includes a restore
service 134
configured to restore the computing platform 102 to a previously saved state.
The
restore service 134 may use snapshots retrieved from the repository of
snapshots 138
to restore the computing platform 102 to a previous state, for example, such
as to a
state saved prior to an erroneous installation of the computing platform 102.
In some
embodiments, the restore service 134 may execute a series of operations
selected to
undo configuration changes made to networking, storage, computing, and virtual
resources. For example, the restore service 134 may instruct a network manager
of the
network 114 to remove a VLAN that was added during installation, or instruct
the
infrastructure manager 112 to de-allocate a blade 116 or storage volume that
were
provisioned during installation.
[0037] In some embodiments, prior to the installation of the computing
platform 102,
an application having one or more functional components may be installed or
loaded
onto the computing platform 102. Each functional component of the application
performs one or more tasks of the application and/or provide a functional
layer of the
application (e.g., in a multi-tiered application). The functional components
include a
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
variety of software components, operating systems, and configurations (e.g.,
executing
on a VM 140) that inter-operate to function as a multi-tiered application. For
example,
the functional components of a deployed web application may include a web
server,
application server, and database server, each executing on a VM 140 from the
virtualized infrastructure 120.
[0038] In some embodiments, an installation process for the computing
platform 102
into an existing data center 100 includes configuring resources of the
computing
platform 102 to enable the pre-loaded application and its functional
components that are
running within the computing platform 102 to communicate with existing
services 106 of
the data center 100. In one example, the platform manager 130 may make
configuration changes to the networking, storage, computing, and virtual
resources of
the computing platform 102 for a virtual desktop infrastructure (VD!) running
within the
computing platform 102, and is shown in greater detail in Figure 2.
[0039] As shown, the platform manager 130 also includes a resource
discovery
service 133 configured to identify resources (e.g., servers 104, services 106)
that are of
interest to the system administrator 150 seeking to connect the newly-deployed
computing platform 102 with the rest of the data center 100. In some
embodiments, the
resource discovery service 133 may identify network switches and blades of
network
108 that have to be configured to allow communication between the computing
platform
102 and the rest of existing data center 100.
[0040] In one embodiment, the platform manager 130 is configured to
generate an
infrastructure template 139 based on the resources identified by the resource
discovery
service 133. The infrastructure template 139 includes infrastructure
parameters that
describe the existing computing environment (e.g., data center 100) in which
the
integrated computing platform 102 is being deployed. Examples of
infrastructure
parameters specified by the infrastructure template 139 include: VLAN
identifiers that
the network 114 of the computing platform 102 should use for communicating
with the
network 108 of the existing data center 100, IP addresses of a DNS service
running
within the data center 100, and a range of IP addresses, subnet masks, and
gateway IP
address to be assigned to VMs 140 of the computing platform 102. In some
embodiments, the infrastructure template 139 may be generated specific to an
application deployed within the computing platform 102. As such, the
infrastructure
11
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
parameters contained in the infrastructure template 139 may address specific
configurations, settings, and information needed by the application executing
within the
computing platform 102. Infrastructure template 139 may be implemented using
any
suitable structured or semi-structured data structure, such as an Extensible
Markup
Language (XML) document, relational database, and key-value data store.
[0041] In some embodiments, the platform manager 130 may be configured
to import
and export infrastructure templates 139 allowing them to be transferred
between other
instances of converged infrastructure (e.g., computing platform 102). The
platform
manager 130 may also import and export of infrastructure templates 139 to
backup
known configurations and settings of the computing platform 102, test and
deploy a
known infrastructure template 139, and perform other such administrative
tasks.
[0042] As shown, the platform manager 130 also includes a resource
configuration
service 135 for configuring physical resources and virtual resources of the
computing
platform 102 based on an infrastructure template 139. For example, the
resource
configuration service 135 may configure the network switches and blades based
on the
infrastructure template 139 to allow communication between the converged
infrastructure (e.g., computing platform 102) and the existing data center
100. In
another example, the resource configuration service 135 may configure network
interface cards (NICs) of VMs 140 running within the computing platform 102
based on
infrastructure parameters of the infrastructure template 139.
[0043] Figure 2 illustrates a computing platform 102 executing an
application 200 to
be integrated into the existing data center 100, according to one embodiment
of the
present disclosure. As shown, the platform manager 130 may deploy a virtual
desktop
infrastructure (VD!) 200 on a plurality of VMs 140 within the virtualized
infrastructure
120. One example of VD! system 200 includes the VMware View system available
from
VMware, Inc.
[0044] In a VD! system 200, an end user 210 uses a VD! client software
program
(e.g., VD! client 212), running on an operating system of a local computing
device, to
access their desktop which may be running in one of VMs 140 in the computing
platform
102 that may be remote from the end user's location. Note that the term
"desktop"
generally refers to the instance of an interactive operating environment
provided by a
12
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
computer operating system and software applications, typically in the form of
a display
and sound output and keyboard and mouse input. With VD! clients 212, users can
access remote desktops 206 running in a remote data center (e.g., computing
platform
102) through the network 128, from any location, using a general purpose
computer
running a commodity operating system and a VD! client software program such as
VMware ViewTM, or a special purpose thin client such as those available from
Dell, HP,
NEC, Sun Microsystems, Wyse, and others.
[0045] As shown, the VD! system 200 includes a connection server 202
that provides
user authentication for remote desktops 206 and direct incoming desktop
requests (e.g.,
from VD! client 212) to a corresponding remote desktop 206. Illustratively,
the VD!
system 200 further includes one or more security servers 204 (e.g., executing
in one or
more VMs 140) that enable secure access to the remote desktops 206 from an
external
network, such as the Internet. The security server 204 may act as a proxy host
for
connections inside a trust network (e.g., network 114) and shields the
connection server
202 from request a public-facing Internet. For simplicity, a single network is
shown but it
should be recognized that, in actual implementations, the components of VD!
system
200 may be connected over the same network or different networks. Furthermore,
a
particular configuration of the virtualized desktop infrastructure is
described above and
illustrated in Figure 2, but it should be recognized that one or more
embodiments of the
present invention may be practiced with other configurations of the
virtualized desktop
infrastructure.
[0046] During installation of the VD! system 200 and the computing
platform 102
within the data center 100, the connection server 202 may be connected to a
domain
controller 208, such as Microsoft Active Directory , already running within
the existing
data center 100 (e.g., on a server 1042). The domain controller 208 manages
user
accounts 214 (e.g., a user account for end user 210) including user log-in
information
and credentials. Further, the connection server 202 and security server 204
may be
connected to a domain name system (DNS) service 216 which is installed on a
server
1041 outside of the computing platform 102 to provide domain names to the
functional
components of the VD! system 200 (e.g., connection server 202, security server
204,
and remote desktops 206). The virtualization manager 122 may need to be
connected
with a "management plane" within the data center 100 and be able to access the
13
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
connection server 202 and security server 204. Network architecture for the
VD! system
200 may be set up that includes one or more virtual local access networks
(VLANs)
between the connection server 202, security server 204, remote desktops 206,
domain
controller 208, DNS service 216 across networks 108 and 114. For example, a
system
administrator 150 needs access to the connection server 202 and security
server 204 to
configure them. Further, a VD! client 212 operated by an end user 210 would
need
access to the connection server 202 and the security server 204 (e.g., via the
networks
128, 108, 114). The one-to-many relationships between the functional
components of
the VD! system 200 and the network 128 of the data center 100 that would need
to be
extended onto the computing platform 102 to ensure proper operation of the VD!
system
200.
[0047] While Figure 2 depicts a particular example of an application
(e.g., VD!
system 200), other applications running within the computing platform 102 may
be
deployed that have components that connect to and communicate with services
106
executing in the data center 100. For example, a computing platform 102 having
an
application with an application server layer, a data grid layer, and a
database layer may
be integrated within the services of the existing data center 100.
[0048] Figure 3 is a flow diagram of method steps for deploying an
integrated
computing platform in an existing data center, according to certain aspects of
the
present disclosure. As shown, the method 300 begins at step 302, where the
platform
manager 130 captures a "factory state" of the computing platform 102. The term
"factory state" may be used to describe a state of the computing platform 102
prior to
the resources of the computing platform 102 being initialized and configured
(e.g., as in
steps 304 and 308 below). In one embodiment, the snapshot service 132 takes an
initial snapshot of the networking resources (e.g., network 114), storage
resources (e.g.,
SAN 118), computing resources (e.g., servers 116), and virtual resources
(e.g., VMs
140) on the computing platform 102. The operations for capturing a state of
the
computing platform are described in greater detail in conjunction with Figure
4.
[0049] At step 304, the platform manager 130 provision resources of the
computing
platform 102 for executing an application and configures the resources to
communicate
with the data center 100. In some embodiments, the platform manager 130
allocates
physical and virtual resources from the computing platform 102 for executing
functional
14
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
components of the application. In one embodiment, the platform manager 130
configures the resources of the computing platform 102 to inter-operate with
services
106 of the data center 100. The platform manager 130 may direct the
infrastructure
manager 112 to configure the networking resources (e.g., network 114) and
hosts (e.g.,
servers 116) to assign VLANs, port groups, set qualify of service (QoS)
settings, and
other parameters need to connect the functional components of the application
to the
services 106 of the data center. The platform manager 130 may direct the
virtualization
manager 122 to configure VMs 140 to add a network interface card (NIC) for
accessing
services 106 outside of the computing platform 102 and for services 106
outside of the
computing platform 102 to communicate with the VMs 140.
[0050] In the example of the VD! system 200, the platform manager 130
may
configure the resources of the computing platform 102 to enable communication
between the connection server 202 to the domain controller 208, to enable
communication between the security server 204 and the network 108, and to
enable
connection between the plurality of remote desktops 206 and the network 108.
Examples of configurations that may be set include assigning an IP address
configured
for management of the domain controller 208, assigning an IP address
configured for
access to the domain controller 208, assigning authentication information for
the domain
controller 208, assigning a VLAN ID that the computing platform 102 should use
for
communicating with the "management plane" of the data center 100, assigning
VLAN ID
that the computing platform 102 should use for communicating with the data
center
"access" network, assigning IP addresses for management and user access to be
assigned to the functional components of the VD! system 200, assigning the IP
address
of the DNS service 216 on the data center network 108, and assigning a subnet
mask
and gateway IP address for the VMs 140 within the computing platform 102.
[0051] In one embodiment, the platform manager 130 may configure the
resources of
the physical infrastructure 110 and virtualized infrastructure 120 based on
values
provided a system administrator 150 (e.g., entered via a graphical user
interface).
[0052] In another embodiment, the platform manager 130 may configure the
resources of the physical infrastructure 110 and virtualized infrastructure
120 based on
an infrastructure template. An infrastructure template specifies
infrastructure
parameters that describe the existing computing environment (e.g., data center
100) in
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
which the integrated computing platform 102 is being deployed. Examples of
infrastructure parameters specified by the infrastructure template may include
VLAN
identifiers that the network 114 of the computing platform 102 should use for
communicating with the network 108 of the existing data center 100, IP
addresses of a
DNS service running within the data center 100, and a range of IP addresses,
subnet
masks, and gateway IP address to be assigned to VMs 140 of the computing
platform
102. In some embodiments, the infrastructure template may be provided for a
specific
application deployed within the computing platform 102. As such, the
infrastructure
parameters contained in the infrastructure template may address specific
configurations,
settings, and information needed by the application executing within the
computing
platform 102. In one implementation, the infrastructure template may be
implemented
as an Extensible Markup Language (XML) document, though any suitable
structured or
semi-structured data structure may be used, such as a relational database or
key-value
data store. An example of an infrastructure template provided for installing a
computing
platform 102 having a deployed VD! system 200 with the data center 100 is
shown in
Table 1 below.
<?xml version="1.0" encoding="UTF-8" ?>
<Infra-Template>
<VLans>
<VLan>
<id>4040</id>
<name>lnfra</name>
</VLan>
<VLan>
<id>4051</id>
<name>Client</name>
</VLan>
</VLans>
<DHCPVirtuaIMachines>
<VMNameFilters>
<VMNameFilter>*CLIENT*</VMNameFilter>
<VMNameFilter>SC-*</VMNameFilter>
</VMNameFilters>
</DHCPVirtualMachines>
<VirtuaIMachinesWithStaticIp>
<VM>
<VMName>VM01</VMName>
<Ip>192.168.15.150</Ip>
<GatewayIp> 192.168.15.1</GatewayIp>
<mask>255.255.255.0</mask>
16
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
<DNS>I92.168.15.150</DNS>
<Domain>example.com</Domain>
</VM>
<VM>
<VMName>VM02</VMName>
<Ip> 192.168.15.151</Ip>
<GatewayIp> 192.168.15.1</GatewayIp>
<mask>255.255.255.0</mask>
<DNS>192.168.15.150</DNS>
<Domain>example.com</Domain>
</VM>
</VirtualMachinesWithStaticIp>
</Infra-Template>
Table 1 - Example Infrastructure Template
[0053] In embodiments where the platform manager 130 configures the
resources of
the computing platform 102 using an infrastructure template, the platform
manager 130
may store the infrastructure template within the repository of snapshots 138,
as
information to be used for any rollback operations in the future.
[0054] At step 306, the snapshot service 132 captures a post-
configuration state of
the computing platform 102. As described in greater detail later, the restore
service 134
may use the captured post-configuration state to restore the computing
platform 102 to
its factory state.
[0055] At step 308, the platform manager 130 installs the application
having a
plurality of functional components using resources of the computing platform
102. For
example, the platform manager 130 may deploy a VD! application (e.g., VD!
system
200) that includes a plurality of VMs 140 executing a connection server 202,
security
server 204, and remote desktops 206. In one embodiment, the platform manager
130
may invoke a call that directs the virtualization manager 122 (e.g., through a
set of APIs)
to create one or more VMs (e.g., VMs 140) having virtual resources (e.g.,
VRAM,
storage) to execute the functional components of the application. The platform
manager
130 deploys instances of the functional components on the allocated resources.
For
example, the platform manager 130 may install software packages onto the
provisioned
VMs 140, or alternatively, the platform manager 130 may invoke a call that
directs the
virtualization manager 122 (e.g., through a set of APIs) to create one or more
VMs (e.g.,
VMs 140) based on a pre-packaged VMs having the application components and
guest
17
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
operating system pre-installed thereon. In some embodiments, the platform
manager
130 may create instances of a functional component based on a VM template that
defines a VM having pre-installed software components, operating system, and
configurations corresponding to a particular functional component.
[0056] At step 310, after the installation of the application has been
completed, the
snapshot service 132 may take a snapshot of the networking, storage,
computing, and
virtual resources of the computing platform 102 that captures the post-
installation state
of the converged infrastructure platform (e.g., computing platform 102). The
post-
installation snapshot may be captured similarly to the snapshots generated in
steps 302
and 306 above.
[0057] At step 312, the platform manager 130 finalizes set up of the
application and
launches the application. In the example of the VD! system 200, the platform
manager
may deploy a pool of remote desktops 206 supported by VMs 140 executing in the
virtualized infrastructure 120. In one embodiment, the platform manager 130
may
import a VM template that serves as a model for the remote desktops 206
(sometimes
referred to as a "gold image") and deploys multiple VMs based on the gold
image to
create a pool of remote desktops for the VD! system 200.
[0058] Figure 4 is a flow diagram of method steps for capturing state of
an integrated
computing platform, according to certain aspects of the present disclosure.
Persons
skilled in the art will understand that, even though the method 400 is
described in
conjunction with the systems of Figures 1 and 2, any system configured to
perform the
method steps, in any order, is within the scope of embodiments of the present
disclosure.
[0059] At step 402, the snapshot service 132 takes a snapshot of network
configurations for network 114 of the computing platform 102. In some
embodiments,
the snapshot service 132 records information regarding VLANs assignments,
logical
networks, port groups, and other network configurations of virtual switches,
IP switches,
Ethernet switches, and storage switches (e.g., fibre channel) that comprise
the
networking resources. The snapshot service 132 may store the snapshot of the
network
configurations in the repository of snapshots 138.
18
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0060] At step 404, the snapshot service 132 takes a snapshot of
computing
resources before configuration of the application is complete. For example,
the
snapshot service 132 records a state of the servers 116 (e.g., blades, server
chassis,
fabric interconnects). At step 406, the snapshot service 132 takes a snapshot
of
storage configuration for the storage resources (e.g., SAN 118). For example,
the
snapshot service 132 records a state of the storage configurations including
storage
arrays, logical volumes, RAID volumes, data replication, backup and recovery
settings.
In some embodiments, the snapshot service 132 communicates with the
infrastructure
manager 112 to obtain configuration states for the computing and storage
resources.
The snapshot service 132 may store the snapshot of the computing and storage
configurations in the repository of snapshots 138.
[0061] At step 408, the snapshot service 132 takes a snapshot of the
state of the
virtualized infrastructure 120 including the plurality of VMs 140. As shown,
the snapshot
service 132 determines, for each of the VMs 140, whether a VM had been cloned
or
instantiated from a VM template. For example, the snapshot service 132 may
determine
that a VM 140 executing connection server 202 may have been derived from VMs
templates for functional components of the VD! system 200. In another example,
the
snapshot service 132 may determine that a VM 140 may have been created from a
VM
template from a library of VM templates provided by the virtualization manager
122.
[0062] At step 410, responsive to determining that the VM was not derived
from a
template, the snapshot service 132 may take a snapshot of the VM 140 and
record the
snapshot in the repository of snapshots 138. In one embodiment, the snapshot
may
include a container file containing serialization of all data, environment,
and state of the
VM 140, for example, formatted in an Open Virtualization Format (OVF) or other
suitable
structure. At step 412, responsive to determining that a VM was derived from a
VM
template, rather than take a snapshot of the VM, the snapshot service 132
records an
association or link between the VM template and the derived VM. The snapshot
service
132 may record the association or link between the derived VM and the VM
template
within the repository of snapshots 138. It should be recognized that steps
408, 410, and
412 may be repeated for capturing snapshots of each of the VMs 140 in the
virtualized
infrastructure 120.
19
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0063] In some embodiments, the snapshots 138 may be implemented as a
recordation of all configuration operations executed on resources of the
computing
platform 102. To capture snapshots, the snapshot service 132 may gather (e.g.,
from
the infrastructure manager and virtualization manager) logs that report what
configuration operations were performed on what resources of the computing
platform
102. The snapshot service 132 may limit what logged operations it collects
based on a
time period. For example, the snapshot service 132 may define a "before"
snapshot as
a point in time prior to configuration of the resources and define an "after"
snapshot as a
point in time after the configuration of the resources (e.g., step 304 above).
The
snapshot service 132 may store this collection of operations in the repository
of
snapshots 138.
[0064] After installation of the application has been completed and the
application is
deployed, a system administrator 150 may later wish to undo the installation
of the
application and restore the computing platform 102 to its factory state (or
other states
during the installation process). For example, the system administrator 150
may
determine that the application is operating improperly and may desire to redo
the
installation process. In another use case, the system administrator 150 may be
experimenting with a variety of different deployment schemes during a proof of
concept
(POC) or testing phase, and wishes to start "fresh" from the beginning of the
installation
process each time.
[0065] Figure 5 is a flow diagram of method 500 for restoring
configuration changes
to an integrated computing platform installed in an existing data center,
according to
certain aspects of the present disclosure. Persons skilled in the art will
understand that,
even though the method 500 is described in conjunction with the systems of
Figures 1
and 2, any system configured to perform the method steps, in any order, is
within the
scope of embodiments of the present disclosure.
[0066] At step 502, the platform manager 130 may receive a "rollback"
command,
e.g., from the system administrator, for restoring configuration changes to
the resources
(e.g., network 114, SAN 118, servers 116, VMs 140, etc.) of the computing
platform
102. In one embodiment, the rollback command may select a snapshot from the
repository of snapshots 138 to which the state of the computing platform 102
is to be
restored. In another embodiment, the rollback command may indicate a subset of
the
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
resources of the computing platform 102 to be restored. For example, the
rollback
command may indicate that only configuration changes to the network 114, and
not to
the storage 118 or VMs 140, are to be restored to the state of the selected
snapshot.
As such, the rollback command may provide a component-by-component granularity
in
its ability to restore configuration state of the resources of the computing
platform 102.
For sake of the foregoing discussion, the rollback command is assumed to
indicate all
levels of resources (e.g., networking, storage, computing, and virtual
machines) have
been selected for restoration.
[0067] At step 504, the restore service 134 determines the changes made
to the
networking, storage, and computing resources based on a comparison of the
snapshots
taken before and after configuration. In one embodiment, the restore service
134
determines an inverse operation that undoes a configuration change between the
before
snapshot and the after snapshot. For example, the restore service 134 may
determine
that a port group was added to the network 114. The restore service 134 may
then
formulate a "delete" operation to delete the port group from the network 114.
In another
example, the restore service 134 may determine that a "delete vlan" operation
is needed
to undo an "add vlan" operation that was performed between snapshots. The
restore
service 134 may determine similar inverse operations for the storage and
computing
resources.
[0068] At step 506, the restore service 134 performs the determined inverse
operation to undo changes to the network, storage, and computing resources of
the
computing platform 102. In one embodiment, the restore service 134
communicates
with the infrastructure manager 112 (e.g., via an API call) to execute the
inverse
operation and restore the network 114, storage 118, and servers 116 to their
factory
state.
[0069] Beginning at step 508, the restore service 134 determines whether
the VMs
140 are derived from a VM template. At step 510, the restore service 134
deletes any
VMs 140 that were derived from a VM template, retrieves the original VM
template, and
clones a new instance of the VM based on the VM template. Alternatively, at
step 512,
responsive to detecting a VM that was not derived from a VM template, the
restore
service 134 retrieve a snapshot of the VM 140 from the repository of snapshots
138.
The restore service 134 directs the virtualization manager 122 to perform a
revert
21
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
operation to return the state of the VM to that of the snapshot retrieved from
the
repository. It should be recognized that steps 508, 510, and 512 may be
repeated for
restoring state of each of the VMs 140 in the virtualized infrastructure 120.
[0070] Figure 6 is a state diagram 600 illustrating a workflow for
resetting state of an
integrated computing platform, according to certain aspects of the present
disclosure.
The state diagram 600 includes states 602, 604, 606, 608 that represent the
different
states of the computing platform 102 during installation of a VD! application
(e.g., VD!
system 200).
[0071] As shown, after the installation process has been completed
(e.g., at state
608), a system administrator 150 may invoke rollback operation that restores
the state
of the computing platform 102 to one of a variety of states captured during
the
installation process. In one embodiment, the rollback operation may restore
the
computing platform 102 to a factory state 602, which undoes any configuration
changes
made during the installation process. For example, a system administrator 150
may
wish to factory reset the computing platform when relocating the computing
platform 102
to a new site (e.g., new data center 100).
[0072] In another embodiment, the rollback operation may revert the
installation of
the VD! application 200 and restore the computing platform 102 to a post-
configuration
state 604, whereby any changes made to the computing platform 102 during
installation
of the functional components of the VD! system 200 are reverted. For example,
restoring to the post-configuration state 604 would remove the instances of
the
connection servers 202 and security servers 204 deployed during the
installation
process. In yet another embodiment, the rollback operation may delete the pool
of
remote desktops and restore the computing platform 102 to a post-installation
state 606,
wherein the VD! application 200 and its functional components have been
installed.
This operation enables a system administrator 150 to replace the "gold image"
or VM
template that defines instances of the remote desktop 206 and deploy a new
pool of
remote desktops.
[0073] Accordingly, embodiments of the present disclosure advantageously
enables
system administrators 150 to easily test their applications and reset the
converged
infrastructure platform to defaults in order to test various use cases. As a
result, the
22
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
time and cost to do a proof of concept (POC) or test deployment is greatly
reduced,
thereby reducing the time to integrate components of a converged
infrastructure with
services of a data center. Embodiments of the present disclosure further
reduce the risk
of error during configuration of network resources and security services,
which may be a
manual and error-prone process.
[0074] As noted above, embodiments presented herein include a packaged
application delivery mechanism to deploy applications having multiple
functional
components. Figure 7 illustrates an example operation for deploying an
application in
the computing platform of Figure 1, according to one embodiment. A system
administrator may provide the platform manager 130 with an application package
136 to
be deployed on the computing platform 100 (e.g., on provisioned physical and
virtual
resources). The application package 700 represents a container of one or more
virtual
machines having pre-installed software components, operating systems, and
configurations that comprise a multi-tiered application. Each VM included in
the
application package 700 may represent a functional component 702 of the
application
that performs tasks of the application or provide a functional layer of the
application
(e.g., in a multi-tiered application). For example, an application package for
a typical
web application may include a first VM having a pre-installed web server,
application
server, and application code for the web application, and another VM having a
pre-
installed database server and configured to be connected to the first VM. The
application package 700 may include files, including data, library, and
metadata files in
a format configured for packaging and distributing virtual machines, such as
the Open
Virtualization Format (OVF).
[0075] In one embodiment, the application package 700 also includes one
or more
models 704 indicating relationships between usage of the physical and virtual
resources
(e.g., VMs 140) and the components 702 of the application. The models 704 may
represent the relationships between resources of the computing platform 100
and
components 702 of the application using mathematical equations. For example, a
model 704 may indicate a relationship between a number of instances (e.g., X)
of a
particular component of the application and a number of expected users (e.g.,
Y) of the
application. For example, the model 704 may indicate that for a deployment to
support
Y users of the application, the deployment should include X instances of a
component,
23
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
where X=N+1, and N=(Y/500). Thus, a model 704 for an application may
incorporate
"best practices" for deploying the application. The particular "best
practices" in a given
case may be pre-determined, e.g., according to testing and empirical results
determined
by system engineers or may reflect industry-wide knowledge. In one embodiment,
the
models 704 may further include a detailed listing of physical resources,
virtual
resources, settings, and configurations typically needed to deploy an instance
of the
application. Examples of settings and configurations provided by the models
704
include networking settings, such as network configurations that indicate
which
components 702 may be located in a "management" portion of the network, in an
externally accessible portion of the network, in a DMZ portion of the network,
etc.
[0076] As described in detail below, the platform manager 130 is
configured to adjust
the deployment of the application based on the one or more models 704 and
further
based on deployment parameters 706 provided by a system administrator. For
example, the platform manager 130 may adjust the deployment of a web
application by
increasing the number of VMs having a pre-installed web server based on a
model 704
that indicates an increased number of web server VMs should be deployed based
on
input from the system administrator indicating the expected amount of traffic
for the web
application. The system administrator may provide parameters 706 indicating
deployment details specific to the instance of the application being deployed.
Examples
of parameters 706 include an anticipated number of users (e.g., 500 users),
volume of
traffic (e.g., 200 requests/second; 500 MB/sec), a desired uptime percentage
(e.g.,
99.999%), and a target resource utilization (e.g., 75% CPU utilization).
[0077] After determining whether (and how) to adjust the deployment of
application
based on the model 704 and parameters 706, the platform manager 130 may
provision
virtual and physical resources (e.g., VMs 140-1, 140-2, 140-3, etc.) from
computing
platform 100 and instantiate individual components 202 of the application
package 700
to deploy the application 710 to virtualized infrastructure 120. While Figure
7 depicts
individual components 702 as modules executing on provisioned VMs (e.g., VMs
140-1,
140-2, 140-3, etc.), the platform manager 130 may alternatively instantiate
complete
VMs having pre-installed software components and operating systems based on
information specified by the application package 700. The operations of the
platform
manager 130 are described in greater detail in conjunction with Figure 8.
24
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0078] Figure 8 is a flow diagram of a method 800 for deploying an
application on
computer resources in a virtualized environment, according to one embodiment.
As
shown, the method 800 begins at step 802, where the platform manager 130
receives
an application package having one or more functional components (e.g.,
components
702) for the application to be deployed within the computing platform 100. For
example,
a system administrator may provide an application package for a VD! system to
the
platform manager 130. The application package may contain multiple pre-
packaged
VMs corresponding to each functional component of the application. For
example, an
application package for a VD! system may contain a pre-configured VM having
pre-
installed application components, software, libraries, and guest operating
system for
running a connection server, a pre-configured VM having pre-installed
application
components, software, libraries, and guest operating system for running a
security
server, and a pre-configured VM having pre-installed application components,
software,
libraries, and guest operating system for running a remote desktop.
[0079] At step 804, the platform manager 130 discovers what physical and
virtual
resources are available for deploying the application in the computing
platform 100. The
platform manager 130 may process the application package and determine
resource
requirements (e.g., memory, compute, storage, networking) specified for
executing the
application, and then query the virtualized infrastructure 120 and physical
infrastructure
110 to determine whether such resources are available. For example, the
platform
manager 130 may determine that a VM corresponding to a functional component of
the
application requires 4 GB of RAM, a 64-bit processor, at least 200 GB of
storage, and at
least a gigabit Ethernet connection. The platform manager 130 may then
communicate
with the infrastructure manager 112 and the virtualization manager 122 to
determine
whether such resources are available from the physical infrastructure 110 and
the
virtualized infrastructure 120, respectively.
[ono] At step 806, the platform manager 130 receives deployment
parameters (e.g.,
parameters 706) specifying how to adjust the deployment of the application
using
models 704. The models 704 indicate how the resources provided by the
computing
platform 100 may be used by, for example, the functional components specified
by the
application package 132. The models 704 may indicate a function relating the
use of
physical resources and virtual resources (e.g., VMs) to a number of instances
of each of
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
the functional components. In one embodiment, the model may specify resources
of the
computing platform 100 to be used as a function of deployment parameters
(e.g., as
provided by the system administrator). For example, a model may specify that a
deployment of a VD! system (e.g., VD! system 300) should include X instances
of a
connection server, where X=N+1 and N=(Y/500), in order to support Y users of
the VD!
system. As such, in this example, the system administrator provides a value of
"1000"
for the deployment parameter of "number of users" (e.g., Y) to the platform
manager 130
to adjust the deployment of the VD! system.
[0081] At step 808, the platform manager 130 determines what physical
and virtual
resources to allocate to the deployment based on a model 704 for the
application.
Continuing the VD! example described above, the platform manager 130 may
evaluate
the model 704 using the deployment parameters of "500 users" to determine that
the
deployment should have three instances of a connection server (e.g.,
(1000/500)+1=3).
The platform manager 130 determines an amount of virtual resources (e.g., VMs
140)
sufficient to support execute of the three instances of the connection server.
[0082] At step 810, the platform manager 130 allocates physical and
virtual
resources from the computing platform 100. In one embodiment, the platform
manager
130 may invoke a call that directs the virtualization manager 122 (e.g.,
through a set of
APIs) to create one or more VMs (e.g., VMs 140) having virtual resources
(e.g., VRAM,
storage) to execute one of the functional components of the application. At
step 812,
the platform manager 130 deploys instances of the functional components 702 on
the
allocated resources. In embodiments where the functional components 702 are
software packages, e.g., the platform manager 130 installs the software
packages onto
the provisioned VMs 140. Alternatively, in embodiments where the functional
components 702 of the application package 700 are pre-packaged VMs, the
platform
manager 130 may invoke a call that directs the virtualization manager 122
(e.g., through
a set of APIs) to create one or more VMs (e.g., VMs 140) as clone of the pre-
packaged
VMs having the application components and guest operating system pre-installed
thereon. For example, the platform manager 130 may create three instances of
the
connection server VM, one instance of the security server VM, and 50 instances
of the
remote desktop VM for a deployment of a VD! system.
26
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0083] At step 814, the platform manager 130 configures the allocated
resources and
the deployed instances of the functional components as specified by the model
138. In
one embodiment, the platform manager 130 may "tune" the allocated resources of
the
computing platform 100 to meet the functional requirements of the application.
For
example, the platform manager 130 may modify default networking settings of
the
allocate resources to meet requirements of a VD! system 300 as specified by
the model.
Specifically, the platform manager 130 may configure the network 114 to create
VLANs
for a DMZ, for internal data network, for external access (e.g., via limited
ports of a
firewall), and for management purposes, where each VLAN may have a particular
quality of service (QoS) setting. The platform manager 130 may further
configure the
deployed instances of the functional components 702 of the application based
on
adjustments made according to a model 704. For example, referring to the VD!
example above, an instance of the connection server may have default
configurations
that assume the instance is the sole connection server within the VD! system.
However,
in deployments where multiple connection servers are instantiated (e.g., based
on the
model 704), the platform manager 130 may modify the default configurations of
multiple
instances of the connection server to function together (e.g., for replication
and load
balancing purposes). For example, the platform manager 130 may elect one of
the
instances of the connection server as a "master" instance and enable a
"replica" mode
within the other instances, provide each instance with the known IP addresses
of the
other instances, etc.
[0084] Figure 9 is a flow diagram of a method 900 for integrating a
computing
platform having an application executing thereon into an existing computing
environment, according to one embodiment. As shown, the method 900 begins at
step
902, where the platform manager 130 deploys an application having one or more
functional components using resources of the converged infrastructure (e.g.,
computing
platform 102). Each functional component of the application performs one or
more
tasks of the application and/or provide a functional layer of the application
(e.g., in a
multi-tiered application). The functional components include a variety of
software
components, operating systems, and configurations (e.g., executing on a VM
140) that
inter-operate to function as a multi-tiered application. For example, the
functional
components of a deployed web application may include a web server, application
27
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
server, and database server, each executing on a VM 140 from the virtualized
infrastructure 120.
[0085] In some embodiments, the platform manager 130 allocates physical
and
virtual resources from the computing platform 102 for executing the functional
components. In one embodiment, the platform manager 130 may invoke a call that
directs the virtualization manager 122 (e.g., through a set of APIs) to create
one or more
VMs (e.g., VMs 140) having virtual resources (e.g., VRAM, storage) to execute
the
functional components of the application. The platform manager 130 deploys
instances
of the functional components on the allocated resources. For example, the
platform
manager 130 may install software packages onto the provisioned VMs 140, or
alternatively, the platform manager 130 may invoke a call that directs the
virtualization
manager 122 (e.g., through a set of APIs) to create one or more VMs (e.g., VMs
140)
based on a pre-packaged VMs having the application components and guest
operating
system pre-installed thereon.
[0086] At step 904, the platform manager 130 determines the resources of
the
converged infrastructure (e.g., computing platform 102) that are to be
connected to
components of the existing computing environment (e.g., data center 100). In
some
embodiments, the platform manager 130 may receive input, e.g., from the system
administrator 150, that identifies which resources of the converged
infrastructure have to
be configured and what configuration information the identified resources need
(e.g.,
infrastructure parameters). The system administrator 150 may indicate to the
platform
manager 130 which VMs inside the converged infrastructure need to be accessed
from
outside of the converged infrastructure. The system administrator 150 may
further
indicate which networking components (e.g., network switches, hosts) need to
be
configured to allow integrated of the converged infrastructure 102 with the
data center
100. For example, a system administrator 150 may provide input to the platform
manager 130 that identifies a VM (e.g., "VM01") running a web server as part
of an
application executing in the computing platform 102. In this example, the
system
administrator 150 indicates the web server needs to be configured with at
least one port
with network connectivity (e.g., via a particular VLAN) to a public Internet
for receiving
web requests.
28
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
[0087] At step 906, the platform manager 130 generates a precursor to an
infrastructure template that specifies the infrastructure parameters needed
for the
converged infrastructure to connect to the existing computing environment 100.
The
precursor to the infrastructure template may be a "blank" infrastructure
template that
names which parameters are needed but lacks the corresponding values. In some
embodiments, the precursor to the infrastructure template may be pre-
determined, and
may be pre-generated for a specific application deployed within the computing
platform
102.
[0088] At step 908, the platform manager 130 determines values for the
infrastructure parameters contained in the infrastructure template 138. In
some
embodiments, the platform manager 130 may process the blank infrastructure
template
(e.g., generated at step 906) to determine what infrastructure parameters need
to be
determined. The platform manager 130 may prompt the system administrator 150
(e.g.,
via a step-by-step graphical "wizard") for values to the infrastructure
parameters. In
some embodiments, the platform manager 130 may derive values for some
infrastructure parameters based on values of other infrastructure parameters
received
from the system administrator. For example, the platform manager 130 may
deduce an
"N+1" domain naming scheme based on infrastructure parameter values (e.g.,
"VM01.example.com," "VM02.example.com") for servers 104 within the existing
data
center 100.
[0089] At step 910, the platform manager 130 generates an infrastructure
template
138 that describe s the environment of the data center 100 using the
determined
infrastructure parameters and their corresponding values. At step 912, the
platform
manager 130 configures the resources (e.g., network 114, servers 116, storage
118) of
the physical infrastructure 110 based on the infrastructure template 139 to be
able to
communicate with the data center 100. At step 914, the platform manager 130
configures the resources (e.g., VMs 140) of the virtualized infrastructure 120
based on
the infrastructure template 139 to be able to communicate with one or more
services
106 of the data center 100. At step 916, the platform manager 130 may
optionally
export the generated infrastructure template 139 for later use, as described
above.
[0090] In some embodiments, an infrastructure template 139 may be
configured for
integrating a computing platform 102 into an existing data center 100. That
is, an
29
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
infrastructure template 139 may be configured for a particular application and
associated functional components running within the computing platform 102. As
such,
the platform manager 130 determines values for infrastructure parameters
(e.g.,
configurations, settings, and information) specific to the application
executing within the
computing platform 102. In one example, an application-dependent
infrastructure
template 139 may be generated for a virtual desktop infrastructure (VD!)
running within
the computing platform 102.
[0091] Figure 10 illustrates an example workflow for integrating the
computing
platform of Figure 2 within an existing data center 100, according to one
embodiment of
the present disclosure. As shown, the resource discovery service 133
identifies
resources (e.g., servers 104) that are of interest to the VD! system 200. For
example,
the resource discovery service 133 may identify the existing data center 100
having at
least a domain controller 208, a DNS service 216, and a network 108. In some
embodiments, the resource discovery service 133 discovers resources the VMs
140 that
host the functional components of the VD! system 200 that need to be access
from
outside of the computing platform 102. The resource discovery service 133
further
discovers the networking components (e.g., network switches, hosts) that need
to be
configured to allow integration of the computing platform 102 with the data
center 100.
[0092] In one embodiment, the platform manager 130 may use a
predetermined
blank infrastructure template associated with the VD! system 200 for
integrating the
computing platform 102 with the data center 100. The infrastructure template
139
identifies configurations, settings, and other set-up anticipated to be needed
by the
functional components of the VD! system 200. In some embodiments, the
infrastructure
template 139 for an instance of VD! system 200 may contain configurations for
properly
connecting a connection server to the domain controller 208, for properly
connecting a
security server to the network 108, and for properly connecting a plurality of
remote
desktops to the network 108. Examples of infrastructure parameters specified
by the
infrastructure template 139 include an IP address configured for management of
the
domain controller 208, an IP address configured for access to the domain
controller 208,
authentication information for the domain controller 208, a VLAN ID that the
computing
platform 102 should use to communicate with a management plane of the data
center
100, a VLAN ID that the computing platform 102 should use for communicating
with the
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
data center access network, a range of IP addresses for management and user
access
to be assigned to the functional components of the VD! system 200, the IP
address of
the DNS service 216 on the data center network 108, and a subnet mask and
gateway
IP address for the VMs 140 within the computing platform 102.
[0093] As shown, the system administrator 150 provides the resource
discovery
service 133 with values for infrastructure parameters 1000 (e.g., via a
graphical user
interface). For example, the system administrator 150, in response to a GUI
query, may
specify the DNS service 216 running in the data center 100 is located at IP
address
"192.168.15.150". In another example, the system administrator 150 may specify
a
VLAN for functional components of the VD! system 200 having a VLAN ID of
"4040" and
a VLAN label of "Infra". In some embodiments, the platform manager 130 may
derive
values for some infrastructure parameters for the VD! system 200 based on
values of
other infrastructure parameters received from the system administrator. For
example,
under certain deployments of a VD! system 200, it may be advantageous to have
user
accounts 214 organized into a specific "organizational unit" for use with VD!
system 200.
As such, using the IP address and authentication information of the domain
controller
network, the platform manager 130 may connect to the domain controller 208 to
determine whether such an organizational unit (OU) already exists, and if not,
to create
one for use with the VD! system 200.
[0094] The platform manager 130 generates the infrastructure template 139
for
integrating the computing platform 102 having the specific application (e.g.,
VD! system
200) executing within. In some embodiments, the generated infrastructure
template 139
may be exported for later re-use. Additionally, an infrastructure template 139
may be
imported from an earlier deployment of an instance of the computing platform
102 and
used to integrate the computing platform 102 within the data center 100.
[0095]
As shown, the resource configuration service 135 uses the infrastructure
template 139 to configure the resources of the computing platform 102 to inter-
operate
with the services 106 of the data center 100. In one embodiment, the resource
configuration service 135 configures the network components (e.g., network
114) and
hosts (e.g., servers 116) to create VLANs using parameters specified in the
infrastructure template 139. In one embodiment, the resource configuration
service 135
adds a network interface card (NIC) to VMs 140 that have need to be accessed
from
31
CA 02882751 2015-02-20
WO 2014/036135
PCT/US2013/057075
outside of the computing platform 102 and configures the NICs using parameters
specified in the resource configuration service 135.
[0096] Various embodiments of the present disclosure may be implemented
as a
program product for use with a computer system. The program(s) of the program
product define functions of the embodiments (including the methods described
herein)
and can be contained on a variety of computer-readable storage media.
Illustrative
computer-readable storage media include, but are not limited to: (i) non-
writable storage
media (e.g., read-only memory devices within a computer such as CD-ROM disks
readable by a CD-ROM drive, flash memory, ROM chips or any type of solid-state
non-
volatile semiconductor memory) on which information is permanently stored; and
(ii)
writable storage media (e.g., hard-disk drive, a USB flash memory device,
etc.) on which
alterable information is stored.
[0097] The invention has been described above with reference to specific
embodiments and numerous specific details are set forth to provide a more
thorough
understanding of the invention. Persons skilled in the art, however, will
understand that
various modifications and changes may be made thereto without departing from
the
broader spirit and scope of the invention. The foregoing description and
drawings are,
accordingly, to be regarded in an illustrative rather than a restrictive
sense.
[0098] While the foregoing is directed to embodiments of the present
disclosure,
other and further embodiments of the present disclosure may be devised without
departing from the basic scope thereof, and the scope thereof is determined by
the
claims that follow.
32
