How did AT&T manage the technology–business–policy framework?
AT&T, ECOMP Policy Framework: Capturing Network
Intelligence in Policies:
ECOMP is accountable for the look, creation, and life-cycle
management of virtualized network functions (VNFs). in a veryll|one
amongst|one in every of} its goals is to try and do this in a
versatile, dynamic, policy-driven manner – belongings users
dynamically management ECOMP’s behavior while not dynamical the
system package. ECOMP’s policy element permits North American
countries to specific, interpret and assess policies, so pass them
on to alternative ECOMP elements or network components for social
control.
Policies manage numerous aspects of ECOMP’s behavior,
including:
1.Service style example: constraints concerning wherever to put
VNFs
2.VNF modification management example: however package rollouts
ought to be scheduled across VNFs and what health checks ought to
be performed to validate changes
3. Management of the behavior of ECOMP elements – e.g., however and
once ECOMP ought to collect information, and the way long ECOMP
ought to retain that information
ECOMP’s policy framework consists of the subsequent
modules:
Policy creation: Enables policies to be specified
by services designers and network operators via computer program
and arthropod genus
Policy evaluation: Evaluates/executes applicable
policies reactively (response to a query) and proactively
(evaluates a policy and triggers actions)
Policy call distribution: Distributes policy
selections to be implemented in alternative ECOMP elements
Policy validation: Validates policies, designed to
attenuate the danger of introducing unhealthy policies into the
ECOMP platform.
The policy framework empowers operators and designers to manage
ECOMP’s behavior. However, the domain information captured in ECOMP
policies is usually distributed across several network operators
and engineers, creating it a challenge to collate. Machine learning
involves the rescue here by serving to service suppliers
mechanically learn policies. within the case of management loops,
machine learning permits signatures and responses to be
mechanically captured.
When the software-centric network effort began, we have a tendency
to had already been adopting cloud technology, totally on IT or
company applications. Cloud technology provided the power to manage
numerous workloads of applications dynamically,
including:
1.Creating Virtual Machines (VMs) on hardware in close to period of
time
2.Assigning applications and workloads to VMs in close to period of
time
3.Dynamically moving applications and dependent functions to
totally different VMs across numerous locations
4.Dynamically dominant resources on the market to applications
(CPU, memory, storage).
AT&T has outlined 3 key objectives across our network
virtualization initiatives:
1.Improve cycle time
2. Open our network
3.Increase potency
Policy: Modifiable Rules, Assertions and/or
Conditions to change period of time higher cognitive process on
corrective actions and configuration changes within the
software-centric network system.
Thank U:)
Get Answers For Free
Most questions answered within 1 hours.