KATE: Hello everyone and thank you for attending today's webinar Chosing the right orchestration
solution for multi-vendor networks sponsored by Anuta Networks. Before we begin I want
to cover a few housekeeping items on the right-hand side of your screen is the Q and A; and you
have any questions during the webcast you can type your question into the Q and A; box
and submit your questions to our speakers all questions will be saved so if we don't
get to answer you, we may follow up via email at the bottom of your audience console a multiple
application widget you can use. If you have any technical difficulties please
click on the yellow hub widget gave my answers to common questions a copy of the slide deck
is available for download in the green resource widget for the end of today's presentation
we'll ask for your feedback our survey will pop open on your screen and will only take
a minute to complete. The feedback is extremely helpful and online version of the webcast
will available about one day after the event and can be accessed using the same audience
[00:59 to 01:00] early today I'll acknowledge hand over to heavy reading senior analyst
Ross.
ROSS: Great thanks Kate and hello everybody and thanks for joining us today we've got
quite a bit here, so I'm just going to run very briefly through the agenda I will start
off talking about some of the industry challenges and I will be passing it over to Brian who
will talk about network services orchestration and how to implement it then Praveen will
do a product overview of Anuta�s NCX solution. Brian will do a demo and then talk about what
is next for network orchestration then I will wrap it up and open it up for your questions
but please do submit them as we go along. So I'd like to start with a look at some of
the challenges that service providers are dealing with, over on the left you'll see
that they're delivering a range of different types of services, enterprise services include
things like MPLS and increasingly SD-WAN, security services such as firewalls are delivered
to both enterprises and consumers, mobile backhaul is becoming increasingly important
with the explosion of mobile data from smart phones tablets and soon IOT as well.
Video services make up a good chunk of operator traffic both the over-the-top traffic coming
from streaming services like YouTube and Netflix as well as traditional TV over cable and fiber.
Customers have been conditioned to expect rapid service delivery and they want to access
those services anywhere at any time and from any device. In order to accommodate this diversity
and support increasing traffic volumes service providers networks also have to change and
do so more frequently than in the past. All of this leads to increased operational
complexity and the overhead is becoming a burden, operators can no longer afford to
manage their networks manually using proprietary management solutions rather they're looking
for open flexible orchestration platforms to more efficiently manage their operations,
this slide lays out how traditional provisioning compares to network services orchestration
on the left you see that configuring device is one by one leads to delays and service
activation further compounding the issue is a lack of automation in those systems and
lastly because of proprietary management systems each element must be managed separately which
again leads to costly software overhead. On the right you see how network services orchestration
takes a different approach by using service templates which mask the inherent difference
between different vendor devices configuration can be simplified and service activation times
reduced, plus vendor provision processes work equally well against legacy networking devices
virtual networking services and software-defined networking devices. With NSO enterprises can
start automating and orchestrating without having to retrofit their environments or wait
for SDN projects to move through their natural life cycles.
So before I turn over to Brian and we're going to take our first poll question, so we'd like
to know from the audience how familiar are you with network orchestration are you just
learning about it, are you evaluating vendors at this point are you already in production
or do you think that network orchestration is not appropriate for you, so while we wait
for people to answer the question I will throw it out to Brian and Praveen and see where
do you think we are on the curve for network orchestration right now would you imagine
that most people are in the earlier stages or do you think people already well into production
on this. BRIAN: So I think at some level that we're
going to see a pretty interesting mix what I really hope to see in this particular audience
is that they're just learning or evaluating vendor stages we're going to share some insights
I'll help them maybe rapidly select some proper vendors and see where we can help them move
forward on their NSO desires. ROSS: Okay well let's take a look and see
where people are at. Oh it's Joe. Joe pretty early on. Forty-four percent say they're just
learning about network orchestration just about a third are evaluating vendors and actually
got some twenty-two percent are in production and only a handful say that network orchestration
is not for them that's great, so at this point I will hand it over to Brian I will get the
slide over and take it away. BRIAN: Great Thank You Ross and again I�m
Brian Fogg the chief innovation officer here at Technica. I'm pleased to be here with Ross
and Praveen to share our insights and lessons learned around network service orchestration
is a very interesting topic it's very timely and a lot of value can be derived from it
as we'll share those through the rest of the webinar. Starting from the right here Ross
address the adaptability from really two perspectives accommodating or adapting to a first set of
hardware and to more rapidly provision new services through the great feature of templates
and we'll get into some of the specific examples of why those are important to productivity.
On the left-hand side not only does the output increase per person but the nature of the
staff can also change without compromising and quality and quality is really the bridge
between productivity and adaptability here with the use of cleverly defined standard
templates a lot of the complexity in hearing and multi-vendor configurations and that is
having different parameters different command sets different definitions can be harmonized
and reduced, so what we wanted to is then talk about what is the net effect of moving
through network services orchestration and you can start with concept of do more with
less or more likely kind of what we see in practices that you're going to really do more
with the same, so as the chart shows on the right the actual number of staff the workload
associated with managing and operating these complex network environments we can see dramatic
reductions in the level of manual effort as they move closer and closer to automated and
orchestrated environments and that allows the ratios the number of servers or number
of devices in the case of Network class of devices fragment to increase over time, we
also tend to see the workload shifting the nature of the skills from traditional strong
network engineering on the Left through operations on the right and so the net effect is that
staff increases the demands and that great curve can increase to the right, that there's
a great our align potential to be able to take the staff and reposition them to support
these kinds of activities. So how does this happen. we look at the set
of desired networks service orchestration features. We really want to take a look at
this from a full life cycle perspective and starting with traditional design then creation,
discover, assure, modify and delete across a really large set of intern two devices and
these kind of environments and have different sets of command parameters, different sets
of user interfaces and we want to make sure that we can try to drive those level of complexities
out we want to use these template based service modeling really against any type of device
and so whether you're a Cisco shop you have aspects of the Juniper framework you have
aspects of HP as you look through you want to be able to cover any of those as you go
through and really provide this from an off-the-shelf support perspective.
So that service provisioning activities the ability to work through topology and discovery
support effort if that's not in your environment today and really try to achieve the FKS kind
of model in a complete clean way but beyond these kind of device management activities
there's also the bottom three bullets there get to full programmability of these environments
and you're going to see later in slide 18 discussion on analytics driven service assurance.
So not only can you help automate which can actually help make the network more intelligent
smarter as it is used and put into place whether you're looking to get through full REST API
integration whether you're looking to adapt comprehensive user defined resource management
techniques that you might already have in place and whether you're really able to pull
us into an existing environment the brownfield fields kind of view or actually put two green
fields which is brand new and assuming that sounds interesting, how do you move forward
in these kind of decision cycles. Well we found that there's really two options
that we tend to consider here as we look at ways to implement and select vendors to support
NSO you can look at the build versus buy and then the open source versus vendor specific
kinds of environments from a build versus buy perspective we think the core focus there
is whether the business of managing the network is your business and that will help decide
whether you want to engage or run this or you're really looking to have it smoothly
automated and orchestrated the breadth and the depth a vendor support needed across your
particular environment. So as you have more and varied types of devices
network service orchestration plays bigger and more interesting dividends over time from
an open source versus vendor selection it's really comes down to a couple of things we've
seen a lot over time that your own support model the level of appetite that you're willing
to pay for community support and how does that work and how do you want to participate
there and how well is that infrastructure, even it's open source how is it fit for purpose
how does it help really match and meet the particular unique needs of your environment
and then the next two there are related the ease of configuration and customization and
then time to market speed to implement, so the level of these kind of complex requirements
grows, you want to have a way to manage that so you can get the right set of staff doing
the right set of activities when you need them and then they both share the build versus
by open source versus vendor share a real common aspect which is how you think through
the total cost of ownership. So one of the core common considerations that
we tend to try to make sure our customers are thinking through that free does not necessarily
mean no cost and so the initial acquisition and then ongoing operations as a real strong
area to pay attention to when you're running those cost models to compare the next sections
a little insight into the implementation of network service orchestration and what are
some of the platform considerations that we tend to follow and recommend that those folks
that are on the early part of the life cycle as Ross able to see in the first poll question.
These are the characteristics we tend to use starting at the top they're already kind of
hit this a little bit the breadth and depth of those vendor support needed want to be
careful there about the device lock-in we'll get into that a little bit later and there
may be some questions from the audience that get to this kind of point later the need to
do discovery whether that's just an upfront kind of activity to discover what's in your
environment or in your network versus ongoing support be able to scale to support not just
physical devices which is important but also the continued movement towards virtualized
appliances at various tiers of your architecture and then the ability to support legacy proprietary
open and emerging standards interfaces as part of that network service orchestration
platform suite. Scalability to the size of environments that
you have validated service models, so that as the kinds of infrastructure platforms you're
trying to automate rollout that you have a way to test that as you go through its normal
lifecycle and then the last to really get to the ease-of-use kind of question and what
we really like and you'll see this later in several slides so the rule is to really maximizes
the fit against these platform selection considerations for wide variety of customer networks and
we're pretty excited to see how they have provided value into these complex network
environments. From a high level perspective, we tend to
follow four steps through the network service orchestration lifecycle focusing in the first
two at the service design stage then working through the service deployment activities
we'll get some more detail and how this work�s a little bit later and then running it through
the tail end of a service management framework traditionally an ITIL based or ISO based kind
of operational environment and if this is the kind of high-level mapping of how they
lay out in practice, what I'd like to do is give you a sense from a more granular level
how might you see this come into play, so our service development process here was really
a precursor to some of the stuff that we're going to show later on slide twenty six starts
with gathering the requirements submission of the devices and things that you're really
focused on. So it's that data set of features and the
first blue box at the bottom the set of protocols that you need to support and then moving them
through the management framework to interface on those devices, so bringing in that CLI
or API support as it make sense through YANG and NETCONF those kind of modern features
and then allow those device level services really then to push all the way to the right
to the higher level delivery of services that are really on top of those devices and try
to push that into a particularly operationally driven environment with that I'll take a pause
hand off to Ross so we can look through the second poll question.
ROSS: Okay great thanks Brian, so the next question for the audience is which do you
think is the most important criteria for choosing a network orchestration platform is it multi-vendor
support? is it brownfield support? is it the ease of customization? is it on Prem and cloud-based
delivery models or is it scale and I can imagine that we've probably got a range here and again
Brian and Praveen, what menu you're speaking with potential customers what is it that they're
saying is the most important thing. BRIAN: This is Brian I'll say that multi-vendor
support is always front and center on people's minds when we have these discussions as well
as the ease of customization because there are common networking environments every customer
does have unique needs and so they want to see a nice path forward to see themselves
represented on these platforms. ROSS: I can see that. Praveen any thoughts
on that before we look at the what the audience sent.
PRAVEEN: So we see the multi-vendor support as the biggest draw in this equation but beyond
that depending on the customer sometimes if it is an enterprise we are beginning to see
more and more requests for a cloud-based delivery model, so that some of the complexity of deploying
the orchestration software and integrating with the ecosystem itself is something that
can be delivered out of the cloud and then the we can just focus on their automation
and orchestration, so that is something we have been seeing lately also.
ROSS: Okay great well let's take a look and see what the audience says and again by a
long shot it is multi-vendor support then was easy customization which is what you were
saying Brian not so much on the cloud-based delivery just yet but I think that's something
that's certainly emerging I'm actually surprised to see that Brown field was so low because
most of these environments are not greenfield, although I have been hearing that people are
perhaps starting with some new services as opposed trying to retrofit legacy services,
so that's all very interesting hopefully useful to you, so now I will hand over to Praveen
who will talk give you an overview of Anuta�s NCX platform.
PRAVEEN: Hi Everyone this is Praveen Vengalam and one of the co-founder and VP of engineering
at Anuta. I walk you through the direction of NCX platform the domain that it plays in
and the high level architecture and I will also look through why the NCX platform has
been chosen by some of our customers and also some use cases, so right upfront let's take
a look at the domains that we are playing in any large enterprise customers or any large
SP�s and MSP�s telcos that we are working with they have this automation or orchestration
needs that span across multiple of their internal domains, they may have a data center for the
private cloud environment they may have data centers where they're providing services to
their external customers they may have some workloads in the private cloud may be a branch
automation may be a campus automation there could be different end points in their IT
infrastructure that are catering to their internal businesses or external businesses
but ultimately there are lot of automation needs and that the use cases are different
it could be application deployment in the data centers it could be a hybrid cloud environment
it could be L2VPN it could be L3VPN it could be a CPE management it could be behind the
CPE the campus management ultimately there are multiple endpoints that need to be configured
and in some scenarios it will be on each endpoint maybe we have to configure multiple features
on the endpoint. So we get all of this complexity in mind and
the fact that historically this used to be done using traditional scripting or manual
automation and how the orchestration helps you something that Brian has just walk through
and we'll just take a look at it more and more and also the NCX platform has been in
play enough in multiple of these domains in some of our customers and that's where the
richness of the platform comes into play and the whole openness and being able to onboard
multiple vendors being able to service chain any underlying device capability or any infrastructure
for that matter is the center of the platform. And we also have open API that allow us to
go and deploy the software like headless where we can go integrate with any northbound system
or in some cases where customers are interested they can consume pretty much all the capabilities
using the in portal, so in the subsequent slides we'll get into a slightly more detail,
so on this slide we're taking a look at the NCX high-level architecture and as you can
see from here we offer a model driven layered and abstracted approach to service delivery
and at the bottom of the screen we are seeing the infrastructure the physical infrastructure
virtual infrastructure or it will even be SDN controllers and there will be an underlying
power of the systems comes from the fact that we can discover any inventory it could be
the physical inventory or it could be the configuration or services that is already
deployed in the underlying infrastructure. So it could be through API it could be through
CLI it could be through YANG or NETCONF ultimately we have pretty much the real time configuration
and operational and some of the telemetry data that is coming from the underlying infrastructure
and this information is taken into the NCX and we put that through a translation engine
where we convert that to a common model, so that's where the abstraction comes in and
once the underlying device information is converted to a common model now we can take
that information and do service chaining within one device multiple features or it could be
multiple features across multiple endpoints or multiple devices it would not matter.
And that's where the strength of the platform comes in we support both greenfield and brownfield
deployments the fact that we can keep our inventory up-to-date with what's there on
the device makes us enable that, also there are lot of other components in the ecosystem
like IP address management system log, CMDB image management and external analytics and
assurance platforms that provide specialized services maybe syslog collections there could
be a lot of different components that we will have to integrate into the overall or orchestration
flow for example let's say we have to draw an IP address from an Infoblox or a BlueCat
or some other custom IP address management solution that also forms part of the overall
orchestration flow and all of these pieces can be easily integrated into the orchestration
platform and since we are delivering all of these capabilities out of open model driven
platform we also are capable of providing an API pretty much for all the capabilities
of the NCX. these are northbound systems and the same API is used by the NCX�s portal
to deliver the model driven UI to the operation teams and also to the admins and then also
to the end customer if at all they want to consume the services out of the NCX portal
directly and the same REST content effects can be leveraged by the northbound OSS portals
self-service portals to define and deliver services from the northbound systems into
the NCX platform. And then towards in the high-level picture
I mean what we are delivering is a model driven and a layered and abstracted approach where
we bring in like a vendor neutrality extensibility and maintainability of services in the overall
lifecycle of a service, so on the next slide we're going to see how we can provide assurance,
so orchestration and assurance are more intertwined because once we go and deploy a configuration
the next aspect is we will need to verify the configuration is working, so historically
this was probably done by some of the diagnostic tools like show commands or probably ping
a lot of troubleshooting tools that help us do that.
So the first part of assurance is to make sure whatever configuration are the or the
policy that's deployed on the infrastructure it stays deployed so that is where we ensure
that any configuration change that happens on the infrastructure can be recovered easily
that is something that we do by monitoring the device level modifications device configuration
modifications and rolling back the changes if at all they are unwanted and beyond that
we also ensure that whatever configuration has been configured on the device it stays
deployed and it works as expected and once the configuration is deployed and it is working
as expected next days. We will need to look at some of the telemetry
data like operational data and some of the other statistics to make sure that it is performing
all the SLAs are performing as expected and towards this end we have a mechanism to monitor
telemetry data and operational data and write rules against it and this is where the policy
engine within the NCX comes into picture and the policy engine allows rules to be written
on pretty much any of the data that is part of the NCX database, so going back to the
prior discussion whatever information that we are able to collect from the underlying
infrastructure and from the partner ecosystem can be used to go and write these rules and
the policies. And we also have a state machine where we
can tie these events and the data that is coming or collected across multiple end point
into more like a state so if an interface goes down on one device followed by an interface
going down at some other device then perform this action, those kinds of rules on state
machine can be articulated so that we can go on the perform assurance and also we don't
expect ourselves to be the sole collector in some cases there may be a specialized collection
engine like Net Flow collector or there they may be somebody else may be doing path computation
but they may be doing advanced analytics, so we are integrating into some of those platforms
we have an open API both north east west and south that way we kind of provide a holistic
approach to assurance that is something of the like apps can be like assurance apps can
be built within NCX or data can be consumed to a streaming interface from NCX so that
they can be built externally, both are possible. And why NCX?? so one of the biggest thing
is the multi-vendor nature and also the cross domain nature, so customers don't want to
deploy a different software - different automation or orchestration software for the different
domains, so they want a single orchestration solution to cater to their multiple domains
hybrid cloud, public cloud, private cloud, the core of the network and datacenter interconnection
services they just want one tool to go with because this again goes back to the fact that
they have to train their internal develop teams the operation team and one tool if at
all it can perform all of those jobs well that would be a predominant choice.
And that's where we we've been able to shine and also apart from that being able to provide
feature velocity how quickly you can allow the customer to design develop and productize
a service and also common models we have taken great care to build a common models for a
lot of vendors like switching routing load balancing firewalls VPN a lot of those capabilities
are already converted to common models across the very common vendors out there and service
model definition is going to be sitting on top of these common models.
So that feature velocity is pretty good and all of this we should be able to deploy out
of a highly scalable and the availability architecture there is something that is also
very key to the NCX platform and all of this also through a micro services based model
and also open standards-based model like IETF YANG and these are some of the reasons why
NCX has been chosen by our customers, so on this slide let's take a look at one of our
large deployments at Tata communications. Tata communications is providing many services
to their end customers to a self-service portal and NCX is the orchestration platform of choice
here and Tata communications has operations across multiple geographical locations 130
countries in all and NCX is deployed globally the NCX server deployed in a highly available
mode both in the in APAC region and in North American region and they manage pretty much
all of these things centrally where they have an integration into their OSS portal and then
the self-service portal ties into that and then NCX agents get deployed geographically
distributed so that they can take closely located to the network elements.
So in this scenario they're providing the CPE management at this initial production
and as we are going further we're also operationalizing the WAN optimization as a part of the use
case and then we are also going to be orchestrating the virtual WAN optimization and also we're
going to be deploying the virtual WAN optimization as VNFs in the public cloud also in the Amazon
Web Services and at this time and the customer is automating more than eighty five percent
of the CPE configuration and it's only going to get better from here.
And as you can see physical CPE where brownfield discovery is taken care and eighty five percent
configuration and then comes the hybrid CPE where we have the virtual WAN optimization
and then the cloud-based WAN optimization will be coming into this production deployment
also, on the next slide similar use case but a significant amount of orchestration in that
data center in this scenario whenever a tenant is on boarded the customer in this scenario
has to onboard the virtual elements like a virtual Arcsight and virtual Infoblox, so
they have to go and deploy some virtual elements in the datacenter for each of the customer
that's being onboarded and also in this scenario we're looking at a fairly large deployment
and so we also have Cisco and then Juniper providing the virtual branch services here
and the Juniper is controlled through the Juniper CSO and the Cisco is controlled through
the directly to the NCX using their traditional management interfaces.
And on the Juniper front we work with the Juniper CSO to be able to orchestrate the
branch automation the branch device automation but ultimately in this scenario what we're
looking at is a variety of ways to automate, so datacenter elements physical and virtual
and then the CPE elements to the controller are directly managing the CPE node. Let's
go to the next slide, so on this slide we're looking at the NFV use case and the Telstra
is providing a virtual branch solution to their retail customers and the platform of
choice is OpenStack as the as virtual infrastructure manager and the KVM is a hypervisor.
So Telstra has a back office where they on board the KVM they on board some of the virtual
infrastructure manager component and then they ship this equipment to the customer they
have a plug-and-play that comes in and then we go deploy the VNFs on demand and once the
VNFs are deployed we go in the orchestrate the services on top by doing the service chaining
and then the policy on top, so this is an up-and-coming use case this is in a pilot
stage now and this is getting more and more traction in the industry as we speak and it's
also important to note here that the Anuta NCX platform has a built-in VNF manager that
will allow some of the VNF bring up VNF scale out and scale in to come in.
And the use case for here with F5 and this is one of the recent announcements we have
made and the industry case customer has a data center where they're providing DDoS as
a service out of their cloud environment and they have multiple different vendors that
are being orchestrated in this in this flow and as you can see from where some of the
key criteria for this customer is being able to provide the device and topology abstraction
and then configuration auditing and reconciliation and of course service orchestration across
multiple endpoints is always been there. So on this slide we're going to summarize
very much what NCX is about it provides a open standards-based approach to service modeling
device modeling and putting complete control in the hands of the customers if at all they
have a DevOps team or providing out-of-the-box solutions like the L2VPNs, L3VPNs using the
IETF published models to some of the customers and also providing the software across multiple
domains campus branch Wireless mobile as I said it should not matter and all of this
at scale is what they NCX platform is good at and over to Ross.
ROSS: Actually I think we are going to Brian for the demo. Correct?
BRIAN: Yes great thanks Ross and Praveen, so what I wanted to do is give a quick sense
of what the demos might look like around this particular technology set, so what I wanted
to do is give just a definition a little bit around what this might look like, so this
is a pretty common demo environment that we stand up here at Technica we have a full suite
of multi-vendor demonstration engineering labs that we tend to get in front of our customers
or potential customers and we think this is the best way to really highlight the power
of Anuta NCX is to see it in action and I really invite the audience to reach out to
us after the webinar and we can talk through opportunities how to make that happen this
scenario we're showing here is a pretty common kind of framework we're on the left we have
a traditional web server in this case showing up as a desired capability that the web requests
a client on the right hand side is trying to reach.
And so this heterogeneous mix of vendor equipment if you can look across here as well as protocols
management interfaces really is set up to support this auto provisioning in a layered
way to support a variety of infrastructure at the MPLS level and then to push out towards
those edge services and to be able to bring this under all or to make automation orchestration
control to be able to reach that end web service in a fully automated way, so this is an interesting
example here as you look and see the equipment there's virtualized assets there's assets
understand NETCONF and YANG and there's other ones that are just command line or CLI interface
only. As it goes through its particular example
sets and we don't really have enough time today to do this as a full demo but these
tend to be the types of things that would show up in the particular demonstration capabilities
that we show go through a strong set of templates, so that you can see how to build up not just
a device level abstraction through the templates but actually think through the core services
that you're deploying across those sets of devices as you develop those templates what
we found is really great way to reduce the complexity of adding those circuit parameters
into those templates. So as your operators are working and trying
to configure out these devices they're able to query back to the CMDB if they can get
the right actionable data that they need to help them really support standing up the service
quickly but if you remember back in the beginning at high quality, so as we push through being
able to reduce those type in errors or make those operators feel much more effective as
they push and we do like to really also highlight that these workflow status opportunities that
not only do you see these in action but there's direct feedback as the workflows are executed
and pushed through their lifecycle and one of the great features of these kinds of systems
that you have. The transactional semantics around them, so
if a complex workflow is running and there are issues that has a clean rollback path,
so getting back to the last known good state is another great feature set as we push through
and look at network service orchestration and how Anuta is a great technology platform
to build that kind of framework on top of there's some other advanced topics you got
a sense of this already with Praveen talking through analytics based service assurance,
so it's more than just compliance and remediation really make it actionable intelligence to
fine-tune and optimize your network environments there's also other opportunities to extend
not just a network elements in here. But also data center oriented elements and
Technical has a solution that extends this service orchestration capability out to compute
and storage components as well, so you can really get to a rich end-to-end orientation
of delivering these kinds of services to will also include SDN integration, so it's not
an opportunity where we see SDN is being outside of this it's a fully managed aspect of how
this infrastructure can roll forward get it into these pilot and prototyping capabilities
and also think through adding container supports and those kind of more advanced feature sets
in the compute side of the equation. So one to reiterate a couple of quick items
of how you get from design to operation want to start at those service levels that you're
really trying to implement look at the features common things that can flow into other aggregate
templates and specific by the areas that you're in determine that mix of physical and virtual
objects and attributes build a model and test as you heard from Praveen and then really
get that deployment out there rapidly to get it into service and so if you wrapped all
of those kind of capabilities up at the lower right hand corner you end up with a better
faster cheaper environment that also encompasses lower risk because you're driving through
Software Defined policies. You're removing errors from a people orientation
of data entry orientation and really looking to support not only just those validation
but rollback opportunities as these kind of solutions roll forward so we're very excited
about the potential we're excited about what we're seeing happen in the market right now,
I thank you for your attention and I'll hand it back to ross to run through the last few
slides. ROSS: Thank you Brian I think Praveen you
were going to speak to the value props and I will wrap it up at the end.
PRAVEEN: Sure one second so Anuta�s value proposition is like provide being able to
provide the multi-vendor orchestration for pretty much any involved automation need the
customer may have and I think it's just a matter of converging the traditional DevOps
and then IT Ops and cloud Ops into a framework where we can define and deliver services sort
of a single pane of glass and all of this also keeping in mind that it's not just deploying
the configuration adjust also being able to assure the service and an ongoing basis and
that's where the strength of the platform and also the industries moving towards and
that's about it Ross I think any other questions. ROSS: I think just the last thing I would
just add just the fact that by using the network services orchestration you can get the cross
vendor in the cross technology provisioning across both the physical and virtual networks
and just by adopting these policy driven operations that CSP can automate and orchestrate their
network services with greater flexibility and agility than they could before, so I think
with that we will take a look to see what kind of questions have been coming and I've
been seeing this thing pop up, so let's take some questions does the NCX a sure SLA through
tight monitoring and audit please elaborate so maybe I think maybe this is a question
for Praveen PRAVEEN: I 'll take that so yeah we do we
actually monitor the infrastructure we get pretty much the SNMP trap Syslogs and any
of the telemetry data out there a we can absorb that into the system and apart from absorbing
that into the system we also convert it into a model like a vendor neutral representation,
so that's going to allow us to go and write rules on top that can be vendor neutral themselves
and we can go and like have a multiple conditions work and then we can actually go and say of
ones when these conditions meet go and perform action and I think I mean as we stated before
we are not the only collectors out there and in some cases we will have specialized collectors
and we're going to get into those specialized collectors but ultimately we can bring all
of the decision-making process come together in the orchestration platform and no matter
who is the colloector ROSS: Okay thank you let's see can the NCX
Orchestrator control and manage multi-vendor physical network infrastructure via and SDN
controller this person was saying either the either ODL or no space controller.
PRAVEEN: I like that question Ross, so we do and we already have integration into different
SDN controllers we have integration into the Cisco APIC platform the Juniper Contrail the
CSO and then Nuage we have done these integrations already and what typically boils down to the
controllers are going to provide specific capabilities and on top of which we will still
need to go and create differentiated service catalogs and that's where the NCX platform
comes in and also the controllers typically don't orchestrate all the endpoints I mean
they may be able to take care of the switching fabric or the routing fabric but being able
to bring in like multiple vendors like load balancers firewalls proxys IPAM, DNS circuit
management, so they're going to be a lot of other components that need to be brought together
and that's where the orchestration platform is pretty much indispensable.
ROSS: Thank you I've got a question for Brian did you have hands-on experience with Cisco's
NSO and can you compare that at all with the Anuta�s NCX solution.
BRIAN: Absolutely thank you thanks for throwing that question this is a great opportunity
we're really looking across the set of vendors that are out there and being able to put them
into labs and actually run them through their paces and be able to put them across these
network environments is a really great way to prove out what are the capabilities, so
we've worked through and done Blueplanet, Glue Networks, Tail-f. So Tail-f being in
recent or near recent acquisition through Cisco it's a great opportunity to really put
these in here and what we found and when we compared these head-to-head we really liked
all of the rich feature sets that come with the Anuta platform the ease of implementation
running through the workflows and developing them and to really see how these could work
in real enterprise environment and there's also the issues of integrating across other
end-to-end solutions, so just as Praveen was talking about being able to hit those firewalls
load balancers and other equipment in a clean way as a real nice hallmark of the Anuta platform.
ROSS: Thanks for that let's see can you comment on how NCX would integrate with OSS and BSS
platforms I think this is probably a prevention Ryan�
PRAVEEN: I think I'll take that, so we haven't open API pretty much all the device capabilities
are opened up through the better RESTCONF interface and this can be used to pretty much
to get into any northbound system our customers have been able to already integrate this into
their homegrown OSS portal and there's some of the other portals that this has been integrated
in some scenarios is the Servicenow and the BMC remedy is some of these integrations have
been done and typically this is something that the customers do because they pretty
much know what aspects of the NCX need to be integrated into their OSS and for example
that NCX portal also provides significant capability, so not all things are integrated
up not only a few things and the customer select what aspects need to be integrated
but we've been fairly successful in that scenario. BRIAN: And if I may add from a Technica perspective
what's great here is this is really a hallmark of Technica as a company, so working with
our federal government customers being able to come in and do assessments to understand
the state of their current OSS to be able to provide recommendations on path forward
to really engineer an architect where they ought to be going focusing kind of on skating
analogy is we want to make recommendations while on the skate where the puck is going
not where it's at and this is a great opportunity to take not only real-world large-scale VOD
experience but tied into the best of commercial offerings at this point.
ROSS: Okay great thank you for that tip well there's quite a few coming in here let me
take a look here let's see here's one for you Brian I see OTAs mentioned, can you provide
a bit more detail on how the US federal government can use them.
BRIAN: Great thanks what I like about the OTAs or the other transactional authorities
is they're an acquisition mechanism that allows the government to really prove out pilot and
test these capabilities against their specific unique needs, so we're a member of C5 and
seed is on the other success, so these OTA vehicles together really allow these requirements
to come out it's a simple way for the customers to push through the requirements for a white
paper and to allow us to from the commercial industry side really put these solutions in
front of them, so it's a great way look at our website for more information or again
reach out to us and we can provide you more specific detail around how C5 , Csec and other
OTA tend to work. ROSS: Okay thank you let me see if I've got
anything for Praveen this is a pretty straight forward one Praveen what is the largest NCX
production deployment that you have so far. PRAVEEN: We have a few of the large MSPs have
deployed ten thousand there so far and that nobody is expected to come up to twenty thousand
shortly. ROSS: Okay this is an interesting one I think
pardon me how do you ensure the skill sets required to run the NCX this is something
I know that comes up with what this whole move to some different kinds of orchestration
solutions the skill set issue so see if you guys want to answer that one.
PRAVEEN: So the whole model driven approach and the orchestration has picked up quite
a bit in the industry lately OPEN CONFIC and then IETF did a lot of activity happening,
so from the Anuta�s side we have a certification process in the training program that we train
our customers and partners and our partners themselves go and train their end customers
and this is something that we've been able to do successfully.
BRIAN: I would also add that on slide 7 we really got to a chart to talk about the shifting
nature of the skills and so what we see is that traditional strong network engineering
requirements are still there in terms of skill and capabilities they tend to push early in
the life cycle when you're setting up the workflows setting up the templates and working
through that business logic really that model-driven approach that Praveen was talking about and
then that allows really once the templates are operated or need to be operated upon that
nature of that skill set becomes less technical and more operational and we think that's a
great way for our customers to really fine-tune to really optimize their staffing profiles
to put the real strong engineering expertise where they need it and to allow operators
to be more efficient with their time and their capability to deliver these kinds of services.
ROSS: Great thanks for having that Brian actually have another one for you from my experience
getting CLI and more modern interfaces to work together can be problematic how easy
is it to see this the Anuta platform in action. BRIAN: So thanks the slide 26 that we want
to working through the demo that's exactly why we have our demo environment set up and
our engineering labs in place let's put us through the set of phases with a set of equipment
that does bring this right to the forefront, so whether you're looking at things that are
fairly modern or driving back to even some interesting legacy equipment that has just
a CLI the Anuta platform supports all of them pushes through and can run across a diverse
set so we invite you to reach out to us if you'd like to see it in action we're very
excited about it and we can run through some interesting scenarios.
ROSS: Okay here's this is an interesting one as well I think does SDN obviate the need
for orchestration I think that I'm just saying that.
BRIAN: Think that's a great question and the byline I'd have people remember here that
this is really not SDN or orchestration this is SDN and orchestration and so we approach
this from an Orchestrator, orchestrators perspective and SDN is a great capability set and Praveen
did talk about having those interfaces that can push through the routing and switching
fabric but end of the day you want to get these policies implemented and controlled
across a diverse set of equipment, so those heterogeneous environments. So we think it's
really no it does not obviate it in fact we can enhance it and customers can enhance their
enterprise orientation by really bringing these tools to bear on these kind of environments.
ROSS: Praveen you want to I add anything to that question?
PRAVEEN: I think Brain got it and then add that question of time apology I think we're
good Ross. ROSS: Okay great let's give it another minute
or so while we wait I wanted to just remind folks to please do take the survey at the
end like Kate said it's very helpful for us to know what works and it will help us as
we create additional ones let's see I think that is it for the question so unless Praveen
or Brian had anything I think that we'll be ready to wrap up.
BRIAN: There nothing further I just thank people for participating throughout the hour
we're pretty excited about the technology and the capability here and invite you to
reach out if there's anything that intrigues you.
ROSS: Okay well thanks again everybody I hope you find it useful and I hope to see you on
another webinar soon thanks very much everybody have a good day.
PRAVEEN: Thanks a lot.
For more infomation >> WATCH LIVE: The National for November 5, 2018 — U.S. Midterms, Veterans Accounting, Credit Insurance - Duration: 1:04:37. 
For more infomation >> Doing Tax ... Open to suggestion for What to Play - Duration: 0:30. 
For more infomation >> Schools hit deadline to apply for coding after school grant - Duration: 1:47. 
For more infomation >> Montana State Bobcats prepare for final home game, senior day on Saturday - Duration: 0:38. 
For more infomation >> Voting Tuesday? Tips for a smooth Election Day - Duration: 2:43. 
For more infomation >> Albuquerque woman kidnapped, raped and held captive for 15-hours - Duration: 2:04. 
For more infomation >> Webster County prepares for the Midterm election - Duration: 0:56. 
For more infomation >> 23-year-old man sentenced for 2013 murder - Duration: 0:33. 
For more infomation >> ⛹️Cute Little Girl Funny Video 2018😀Funny Status Video For Whatsapp😀 - Duration: 0:51. 
Không có nhận xét nào:
Đăng nhận xét