View all newsletters
Receive our newsletter - data, insights and analysis delivered to you

Q&A: Stamping out human error with data centre automation

CBR sits down with Nigel Oakley, EMEA director, Cloud-building Centre of Excellence, to talk data centre revolution from ground-up.

By Joao Lima

Provider Juniper Networks believes network plays a central role in addressing the critical challenges companies face as a global community.

At an exclusive PoC lab visit held in Amsterdam by Juniper, Oakley explained how businesses will evolve with the next generation data centre.

CBR: How does the high performance data centre help create a better customer experience and drive business?

NO: It is about the next generation data centre. We are seeing data centres where services have been slow to deliver, although we have seen a transformation of IT in terms of automation.

What we have not seen has been the corresponding change or investment in automation for the network. We know of situations where it would take 30 seconds or a minute to spin up a new application in terms of a virtual machine using the IT systems. Then it would take up to two weeks to connect that application.

What we have observed over the last 12 to 18 months with the advance of software defined networking capabilities is the ability to automate that network layer that exists within the data centres, and to extend that IT transformation that has happened in the IT architecture into the network itself and then deliver a completely automated service.

Once customers can do that, they can transform the quality of the experience because around 90% of data centre network outages are caused by human error. If users can move to an automated world, they can deliver a quicker service, more reliably.

Content from our partners
Unlocking growth through hybrid cloud: 5 key takeaways
How businesses can safeguard themselves on the cyber frontline
How hackers’ tactics are evolving in an increasingly complex landscape

CBR: Can you make the service more reliable by using automation?

NO: Yes, absolutely. By automating the network, customers can automate recovery, automate transfer services between data centres, and they can make sure the replication works effectively.

Parts of the MetaFabric architecture gives them that data centre interconnectivity, and gives the very high performance data centre interconnectivity. If a customer applies automation to that as well, he or she can really enable duplication of the services coming from multiple different data centres.

If we look at the Googles and Facebooks of the world, they have done that already, and they are providing the services from many data centres around the world. In addition, if one of the data centres does not respond, service will come from elsewhere.

Automation is a key component on how customers build out that architecture and the capabilities of the switches and the routers they are putting in to be able to deliver that service.

CBR: What is the MetaFabric solution and how can it help data centre companies?

NO: MetaFabric is an architecture. We took the physical capabilities that we have within our routers and switch platforms, allied with virtualisation technologies and management systems, to build an architecture that consumers can feel confident about deploying from both an agility and scalability perspective.

We can map out for a customer using the MetaFabric architecture exactly how they can build a data centre, the components that go with it, and how they can scale that data centre.

What we did with the MetaFabric architecture was recognise challenges that customers in enterprises were facing and build simplified architectures. This gives the fastest response both to data coming from North to South and East to West.

CBR: A recent report found that 45% of UK companies have experienced down time due to natural causes. How important is replication?

NO: We are seeing two different approaches: private data centre replication with the ability to replicate the data between them; and cloud being used as resilience.

Moving the data and workloads between the private data centre and the cloud data centre, is a crucial capability.

CBR: How essential is Juniper’s Amsterdam PoC lab to develop its solutions?

NO: The PoC lab is a key part of the development cycle. At some point, the customer wants to be able to test [the solution], to kick it in. Customers want to break it because until they have an opportunity to see where it breaks or it does not work, they do not feel confident about being able to repair it if something goes wrong.

It allows them to test interoperability between different vendors and Juniper’s equipment, and to have confidence that the solution that is being proposed will work against the use cases that they have defined.

CBR: How is IoT going to impact your business?

NO: It challenges more. There will be more IP addresses, more scalability, more bandwidth, more data centres being able to cope with more capabilities or more demands that will be placed on them.

The investment we are putting into new technologies like the PTX1000, demonstrates the commitment we have to provide the scalability that IoT will throw on demand on infrastructure technologies.

We invested billions in the chip sets that go into the hardware platforms that deliver that scalability.

From our perspective, IoT drives much more demand for many of the capabilities that we deliver today. Our opportunity is to deliver the platform that is going to provide that scalability.

CBR: With so much activity happening in the European colo space, how do you see the industry changing?

NO: One interesting thing we see happening is more geographic demand to deliver services out of local geography. Certainly one of the challenges that emerged on a geographical scale with the whole Snowden debacle.

We are seeing more vendors saying that they want to be sure their data stays in Germany, France or the UK.

We see a growth of data centres where they are providing that capability within a region, rather than internationally.

CBR: Do you believe every company should be a software company?

NO: The companies that are going to succeed in this next platform of delivery and capability are the ones that really have the vision to use new software techniques and new methods of delivering services to customers and build those into the core of the business.

Rather than IT being a reactive part of the company, delivering servers and configuring laptops, they need to be a core asset at the board level, and being able to deliver that differentiation the next generation of companies will need.

Websites in our network
Select and enter your corporate email address Tech Monitor's research, insight and analysis examines the frontiers of digital transformation to help tech leaders navigate the future. Our Changelog newsletter delivers our best work to your inbox every week.
  • CIO
  • CTO
  • CISO
  • CSO
  • CFO
  • CDO
  • CEO
  • Architect Founder
  • MD
  • Director
  • Manager
  • Other
Visit our privacy policy for more information about our services, how New Statesman Media Group may use, process and share your personal data, including information on your rights in respect of your personal data and how you can unsubscribe from future marketing communications. Our services are intended for corporate subscribers and you warrant that the email address submitted is your corporate email address.