There’s always a balance in IT service delivery. Security vs usability, performance vs cost, agility vs stability. Traditional service delivery models have their place but the rapid rise of end user experience innovation often results in digital teams developing new services far quicker the the infrastructure teams can deliver them. This need to rapidly develop and more importantly deliver new services is the route cause of frustration for many teams which is where ‘Bimodal’ IT comes in.
Gartner suggest this two tier model has a traditional slow ‘mode 1’ element that delivers core services and an agile mode 2 that significantly increases IT agility and reduces the time to deliver new services. Quoting Gartner SVP Peter Sondergaard “This capability (Bimodal) allows the IT organisation to respond to the digital divide within their organisation by operating in two modes that are comprehensive and coherent, but deeply different, while exploiting the benefits of both…”
This mode 2 adoption is driving new challenges of managing core delivery network infrastructure, particularly at scale. Add in the additional complexity of reporting and analytics, integration with multiple orchestration systems and support for multi or hybrid clouds and a very relevant question arises about traditional management tools such as Command Center, Insight Center and Control Center. They are all great tools in their own right but a single system that combines all of their functionality would clearly be advantageous. And that is the basis for the new NetScaler 11.1 Management and Analytics System (N-MAS).
N-MAS allows organisations to transform the way networking operations and developers program the infrastructure to create app delivery services. At the heart of this transformation, N-MAS provides Stylebook Engine which allows network teams and developers to compose service delivery configurations using simple expressions that describe the structure of a NetScaler configuration. This in turn provides continuity between development and operations teams, reducing IT admin complexity and significantly accelerating application deployment. Cool.
It’s not all DevOpps though. N-MAS has been developed with feedback from enterprises, service providers and web monsters to support both agile and traditional IT modes. Supporting NetScaler MPX, SDX, VPX and the new container ready CPX along with recently renamed SDWAN across multiple and hybrid clouds, the capabilities are significant as shown in the image below.
So what does this mean in the real world?
N-MAS for traditional workloads simplifies the management of NetScaler instances by providing;
- Centralised configuration using Stylebooks, created and configured across an inventory of NetScaler instances.
- Centralised SSL certification management.
- Automated software upgrades with provision for ‘rolling’ upgrades for HA pairs to eliminate downtime and reduce operational overheads.
- Automated troubleshooting through the collection of data from all NetScaler instances, identifying trends and which individual VIPs (service access points) are failing to deliver.
- Logging and analytics from an application perspective rather than an appliance perspective. (Nice)
- Microsegmentation and trusted networks to create strong boundaries between apps and services. Only the apps that are allowed to communicate with each other are allowed to do so.
- Multiple AppCentric views with read / write privileges to accommodate the different audiences such as developers, network operations and Infosec teams that need to consume NetScaler resources.
- Hybrid cloud support to allow operations staff to manage all NetScaler instances, regardless of their physical location.
N-MAS for microservices and container environments.
Applications drive infrastructure. Using N-MAS, the infrastructure becomes a pool of resources that can be programmed to fit the different types of apps and use cases. This in turn offers;
- Developers the ability to easily program a NetScaler instance as part of the app development process without prior NetScaler configuration knowledge via Stylebooks.
- Pooled resources as the NetScaler instances are abstracted, allowing them to become part of a pool of resources available for any application. This in turn supports the ability to move any app to any available NetScaler, spin up or scale NetScaler instance on demand and spinning down as required. Super agile!
- Trusted networks and microsegmentation, to separate pre-production, development and production traffic, reducing security risk and internal app to app DDoS threat.
- Service discovery, allowing applications spun up in dynamic environments to find and communicate with each other effectively to support business process and workflows.
- Logging and data management of large numbers of instances from MPX, VPX, SDX and CPX appliances. Up to 1000 instances at launch and 10,000 scheduled H216.
- Analysis of performance metrics for developers, architects and business managers.
- Troubleshooting with a high degree of automation to allow network operators to manage a large number of applications and NetScaler instances.
Command Center, Insight Center and Control Center are to remain for the time being but with new functionality reserved for N-MAS, Citrix have thrown a curve ball in to the market place. No other ADC vendor offers this level of orchestration, management and analytics capability in a single platform. Far greater than the sum of its parts, N-MAS will have price tag for usage when it’s launched in Q2 but like many things IT related, the cost of the alternatives will be far greater in money, time or complexity.
© Al Taylor – cloudDNA
24th May 2016