Tuesday, 22 November 2016

Russia Blocks LinkedIn: A Sign of Things to Come?

It's official. After months of threatening LinkedIn with a block of its services in Russia, the Russian communications regulator has followed through. It all stems from a dispute over whether LinkedIn would comply with Russian laws requiring information pertaining to Russian users be hosted on Russian servers. One wonders if Russia's actions are a sign of things to come.

The push to bring all Russian user data home began in 2014 when the Duma passed the first of several bills aimed at doing just that. Under that first piece of 2014 legislation, Russia established that companies operating on an international scale would have to procure physical servers in Russia – whether contracting with existing data centres or building their own – in order to store data pertaining to Russian users.

The law equally applies to big names such as Facebook and Google and smaller companies with significantly less name recognition. Furthermore, it applies to Russian companies who have a practice of sending user data outside the country. They must cease doing so unless they can prove a certain level of domestic data security.

A Populist Mentality or Something Else?

One way to look at the Russian legislation is to compare it to the current wave of populism that seems to be sweeping the globe. Citizens growing ever more tired of globalism are demanding their nations return to a more populist way of doing things that preserve national identity and sovereignty. Populism was a big part of both Brexit and the recent US presidential election. It may grow in the near future with both the French and German elections.

Could Russia's move be as much about populism as security?  While it's true that protecting sensitive data is a lot easier when hosted domestically, it is also not terribly difficult to implement security strategies that are effective in a cross-border situation. So there has to be more to it than just security alone. Populism seems as if it could be a factor.

Still, there is another possibility. Some critics of Russia's move speculate that the regulator wants data stored at home so that certain government agencies can access it more easily. Think NSA and Edward Snowden here.

Where Do We Go from Here?

Now that Russia has taken steps to block LinkedIn, we would expect the regulator to take similar action against other companies as well. The floodgates are opened and water should begin pouring through rather quickly. Whether it spreads to other countries remains to be seen.

As per LinkedIn, they continue to stay committed to a global mindset. In their official statement, they expressed the following:

"Roskomnadzor's action to block LinkedIn denies access to the millions of members we have in Russia and the companies that use LinkedIn to grow their businesses. We remain interested in a meeting with Roskomnadzor to discuss their data localisation request."

Will Russia flinch? Probably not. So now it's up to LinkedIn to make it work.

Friday, 11 November 2016

Thoughts on a More Predictable & Reliable Data Centre Life Cycle

For a variety of reasons, there has been too much attention paid to the way data centres and infrastructure has been built, with comparatively little attention on the cost of operations throughout the life cycle.

As an industry we are 30-35 years old. We’ve grown very rapidly and so far we’ve been very technology driven and that is an extremely good thing. That focus has created the infrastructure that resides in hundreds of thousands of data centres around the world and it’s that infrastructure that gives us all the things that we take for granted, like the internet and applications such as messaging, streaming, two-way video communications and so on.

The next challenge is quite a different one and it’s making the transition from being engineering-focused to being operationally focused. What that really means is that we need to start to think much more carefully about how all the infrastructure we will have is going to be managed. How is it going to be run? How do we know how well we are running our infrastructure and doing our jobs to the best of our abilities?

Part of this, of course, is people related. But there is also a technical solution which requires giving thought to – what the infrastructure looks like and feels like throughout its life cycle, not just putting a data centre together from a design and build perspective and then moving on to the next project.

We need to think about ourselves as an industry that is maturing and as all industries mature they go through several stages of pain. The initial stages of pain are related to that change, in other words understanding where you are in the process and making the decision to change.
What that means is thinking very carefully about life cycle. How will the infrastructure that is built today perform throughout the phases of its life cycle? At some point in time we will refresh equipment. We will make capital reinvestments. We will make operational investments. We need to think those through throughout the life cycle.

What technology platform do we put in place so that we can manage our infrastructure better? The industry is still in a state of hyper growth so we’re still going to grow the number of facilities, although they may change size and shape. In fact, if the market does change in the way that we expect it to and makes a move towards Edge computing, the whole facility landscape will change dramatically.

To be able to manage the operation of those sites better we need to think about what the life cycle looks like. How do we want to run infrastructure in the best possibly way, ideally with the least amount of human intervention, and that’s where software and technology come in.

One of the things that we can do as an industry is to short-circuit that learning process by not going through the same pains that the other industries have already been through. Let’s look at oil & gas, pharmaceuticals, water and utilities and nuclear power stations. They’ve been through this exercise in varying time periods over the last 10 to 15 years. Let’s figure out what they did to change their operational best practice and use that knowledge.

So we don’t have to learn all those lessons for ourselves; we’ll make our own mistakes and learn our own lessons but let’s at least stand on the shoulders of our ancestors in the context of this particular maturity.

We need to do a better job in two ways. We need to describe our vision of the future and that this vision is not just about products and technology: it is really about the life cycle. Customers no longer buy product; they buy a system. They buy a solution. They buy an entire data centre. We would expect customers to say: “I’d like you to build me infrastructure that is predictable in terms of reliability and efficiency but is also incredibly easy to manage.”

Schneider Electric white paper 195, “Fundamentals of Managing the Data Centre Life Cycle for Owners” describes the five phases of the life cycle, identifies key tasks and pitfalls, and offers practical advice to the owners and management of legacy facilities.

Guest blog by Arun Shenoy, VP, IT & Data Centre Business, Schneider Electric