Friday, February 5, 2021

How storage advances help businesses digitally transform across a hybrid cloud world

The next BriefingsDirect data strategies insights discussion explores how consistent and global storage models can best propel pervasive analytics and support digital business transformation.

Decades of disparate and uncoordinated storage solutions have hindered enterprises’ ability to gain common data services across today’s hybrid cloud, distributed data centers, and burgeoning edge landscapes.

Yet only a comprehensive data storage model that includes all platforms, data types, and deployment architectures will deliver the rapid insights that businesses need.

Stay with us to examine how IBM Storage is leveraging containers and the latest storage advances to deliver the holy grail of inclusive, comprehensive, and actionable storage.

Listen to the podcast. Find it on iTunes. Read a full transcript or download a copy.

To learn more about the future promise of the storage strategies that accelerate digital transformation, please welcome Denis Kennelly, General Manager, IBM Storage. The interview is conducted by Dana Gardner, Principal Analyst at Interarbor Solutions.

Here are some excerpts:


Gardner: Clearly the world is transforming digitally. And hybrid cloud is helping in that transition. But what role specifically does storage play in allowing hybrid cloud to function in a way that bolsters and even accelerates digital transformation?


Kennelly: As you said, the world is undergoing a digital transformation, and that is accelerating in the current climate of a COVID-19 world. And, really, it comes down to having an IT infrastructure that is flexible, agile, has cloud-like attributes, is open, and delivers the economic value that we all need.


That is why we at IBM have a common hybrid cloud strategy. A hybrid cloud approach, we now know, is 2.5 times more economical than a public cloud-only strategy. And why is that? Because as customers transform -- and transform their existing systems – the data and systems sit on-premises for a long time. As you move to the public cloud, the cost of transformation has to overcome other constraints such as data sovereignty and compliance. This is why hybrid cloud is a key enabler.


Hybrid cloud for transformation

Now, underpinning that, the core building block of the hybrid cloud platform, is containers and Kubernetes using our OpenShift technology. That’s the key enabler to the hybrid cloud architecture and how we move applications and data within that environment.


As the customer starts to transform and looks at those applications and workloads as they move to this new world, being able to access the data is critical and being able to keep that access is a really important step in that journey. Integrating storage into that world of containers is therefore a key building block on which we are very focused today.


Storage is where you capture all that state, where all the data is stored. When you think about cloud, hybrid cloud, and containers -- you think stateless. You think about cloud-like economics as you scale up and scale down. Our focus is bridging those two worlds and making sure that they come together seamlessly. To that end, we provide an end-to-end hybrid cloud architecture to help those customers in their transformation journeys.


Gardner: So often in this business, we’re standing on the shoulders of the giants of the past 30 years; the legacy. But sometimes legacy can lead to complexity and becomes a hindrance. What is it about the way storage has evolved up until now that people need to rethink? Why do we need something like containers, which seem like a fairly radical departure?


Kennelly: It comes back to the existing systems. You know, I think storage at the end of the day was all about the applications, the workloads that we ran. It was storage for storage’s sake. You know, we designed applications, we ran applications and servers, and we architected them in a certain fashion.

When you get to a hybrid cloud world ... If you're in a digitally transformed business, you can respond rapidly. Your infrastructure needs to respond to those needs versus having the maximum throughput capacity.

And, of course, they generated data and we wanted access to that data. That’s just how the world happened. When you get to a hybrid cloud world -- I mean, we talk about cloud-like behavior, cloud-like economics – it manifests itself in the ability to respond.


If you’re in a digitally transformed business, you can respond to needs in your supply chain rapidly, maybe to a surge in demand based on certain events. Your infrastructure needs to respond to those needs versus having the maximum throughput capacity that would ever be needed. That’s the benefit cloud has brought to the industry, and why it’s so critically important.


Now, maybe traditionally storage was designed for the worst-case scenario. In this new world, we have to be able to scale up and scale down elastically like we do in these workloads in a cloud-like fashion. That’s what has fundamentally changed and what we need to change in those legacy infrastructures. Then we can deliver more of our analysis-services-consumption-type model to meet the needs of the businesses.


Gardner: And on that economic front, digitally transformed organizations need data very rapidly, and in greater volumes -- with that scalability to easily go up and down. How will the hybrid cloud model supported by containers provide faster data in greater volumes, and with a managed and forecastable economic burden?


Disparate data delivers insights

Kennelly: In a digitally transformed world, data is the raw material to a competitive advantage. Access to data is critical. Based on that data, we can derive insights and unique competitive advantages using artificial intelligence (AI) and other tools. But therein lies the question, right?


When we look at things like AI, a lot of our time and effort is spent on getting access to the data and being able to assemble that data and move it to where it is needed to gain those insights.


Being able to do that rapidly and at a low cost is critical to the storage world. And so that’s what we are very focused on, being able to provide those data services -- to discover and access the data seamlessly. And, as required, we can then move the data very rapidly to build on those insights and deliver competitive advantage to a digitally transformed enterprise.


Denis, in order to have comprehensive data access and rapidly deliver analytics at an affordable cost, the storage needs to run consistently across a wide variety of different environments – bare-metal, virtual machines (VMs), containers -- and then to and from both public and private clouds, as well as the edge.


What is it about the way that IBM is advancing storage that affords this common view, even across that great disparity of environments?


Kennelly: That’s a key design principle for our storage platform, what we call global access or a global file system. We’re going right back to our roots of IBM Research, decades ago where we invented a lot of that technology. And that’s the core of what we’re still talking about today -- to be able to have seamless access across disparate environments.

A key design principle for our storage platform, what we call global access or a global file system, goes back to our roots at IBM Research. We invented a lot of that technology. And that's at the core of what we're talking about -- seamless access across disparate environments.

Access is one issue, right? You can get read-access to the data, but you need to do that at high performance and at scale. At the same time, we are generating data at a phenomenal rate, so you need to scale out the storage infrastructure seamlessly. That’s another critical piece of it. We do that with products or capabilities we have today in things like IBM Spectrum Scale.


But another key design principle in our storage platforms is being to run in all of those environments -- bare-metal servers, to VMs, to containers, and right out to the edge footprints. So we are making sure our storage platform is designed and capable of supporting all of those platforms. It has to run on them and as well as support the data services -- the access services, the mobility services and the like, seamlessly across those environments. That’s what enables the hybrid cloud platform at the core of our transformation strategy.


Gardner: In addition to the focus on the data in production environments, we also should consider the development environment. What does your data vision include across a full life-cycle approach to data, if you will?


Be upfront with data in DevOps

Kennelly: It’s a great point because the business requirements drive the digital transformation strategy. But a lot of these efforts run into inertia when you have to change. The development processes teams within the organization have traditionally done things in a certain way. Now, all of a sudden, they’re building applications for a very different target environment -- this hybrid cloud environment, from the public cloud, to the data center, and right out to the edge.


The economics we’re trying to drive require flexible platforms across the DevOps tool chain so you can innovate very quickly. That’s because digital transformation is all about how quickly you can innovate via such new services. The next question is about the data.


As you develop and build these transformed applications in a modern, DevOps cloud-like development process, you have to integrate your data assets early and make sure you know the data is available – both in that development cycle as well as when you move to production. It’s essential to use things like copy-data-management services to integrate that access into your tool chain in a seamless manner. If you build those applications and ignore the data, then it becomes a shock as you roll it into production.


This is the key issue. A lot of times we can get an application running in one scenario and it looks good, but as you start to extend those services across more environments – and haven’t thought through the data architecture -- a lot of the cracks appear. A lot of the problems happen.


You have to design in the data access upfront in your development process and into your tool chains to make sure that’s part of your core development process.


Gardner: Denis, over the past several years we’ve learned that containers appear to be the gift that keeps on giving. One of the nice things about this storage transition, as you’ve described, is that containers were at first a facet of the development environment.


Developers leveraged containers first to solve many problems for runtimes. So it’s also important to understand the limits that containers had. Stateful and persistent storage hadn’t been part of the earlier container attributes.


How technically have we overcome some of the earlier limits of containers?


Containers create scalable benefits

Kennelly: You’re right, containers have roots in the open-source world. Developers picked up on containers to gain a layer of abstraction. In an operational context, it gives tremendous power because of that abstraction layer. You can quickly scale up and scale down pods and clusters, and you gain cloud-like behaviors very quickly. Even within IBM, we have containerized software and enabled traditional products to have cloud-like behaviors.


We were able to quickly move to a scalable, cloud-like platform very quickly using container technology, which is a tremendous benefit as a developer. We then moved containers to operations to respond to business needs such as when there’s a spike in demand and you need to scale up the environment. Containers are amazing in how quickly and how simple that is.

We have been able to move to a scalable, cloud-like platform very quickly using container technology, which is a tremendous benefit as a developer. We then moved containers to operations to respond to business needs to scale up and down. Containers are amazing in how quickly and how simple that is.


Now, with all of that power and the capability to scale up and scale down workloads, you also have a storage system sitting at the back end that has to respond accordingly. That’s because as you scale up more containers, you generate more input/output (IO) demands. How does the storage system respond?


Well, we have managed to integrate containers into the storage ecosystem. But, as an industry, we have some work to do. The integration of storage with containers is not just the simple IO channel to the storage. It also needs to be able to scale out accordingly, and to be managed. It’s an area we at IBM are focused on working closely with our friends at Red Hat to make sure that’s a very seamless integration and gives you consistent, global behavior.


Gardner: With security and cyber-attacks being so prominent in people’s minds in early 2021, what impacts do we get with a comprehensive data strategy when it comes to security? In the past, we had disparate silos of data. Sometimes, bad things could happen between the cracks.


So as we adopt containers consistently is there an overarching security benefit when it comes to having a common data strategy across all of your data and storage types?


Prevent angles of attack

Kennelly: Yes. It goes back to the hybrid cloud platform and having potentially multiple public clouds, data center workloads, edge workloads, and all of the combinations thereof. The new core is containers, but you know that with applications running across that hybrid environment that we’ve expanded the attack surface beyond the data center.


By expanding the attack surface, unfortunately, we’ve created more opportunities for people to do nefarious things, such as interrupt the applications and get access to the data. But when people attack a system, the cybercriminals are really after the data. Those are the crown jewels of any organization. That’s why this is so critical.


Data protection then requires understanding when somebody is tampering with the data or gaining access to data and doing something nefarious with that data. As we look at our data protection technologies, and as we protect our backups, we can detect if something is out of the ordinary. Integrating that capability into our backups and data protection processes is critical because that’s when we see at a very granular level what’s happening with the data. We can detect if behavioral attributes have changed from incremental backups or over time.


We can also integrate that into business process because, unfortunately, we have to plan for somebody attacking us. It’s really about how quickly we can detect and respond very quickly to get the systems back online. You have to plan for the worst-case scenario.


That’s why we have such a big focus on making sure we can detect in real time when something is happening as the blocks are literally being written to the disk. We can then also unwind to when we seek a good copy. That’s a huge focus for us right now.


Gardner: When you have a comprehensive data infrastructure, can go global and access data across all of these different environments, it seems to me that you have set yourself up for a pervasive analytics capability, which is the gorilla in the room when it comes to digital business transformation. Denis, how does the IBM Storage vision help bring more pervasive and powerful analytics to better drive a digital business?


Climb the AI Ladder

Kennelly: At the end of the day, that’s what this is all about. It’s about transforming businesses, to drive analytics, and provide unique insights that help grow your business and respond to the needs of the marketplace.


It’s all about enabling top-line growth. And that’s only possible when you can have seamless access to the data very quickly to generate insights literally in real time so you can respond accordingly to your customer needs and improve customer satisfaction.


This platform is all about discovering that data to drive the analytics. We have a phrase within IBM, we call it “The AI Ladder.” The first rung on that AI ladder is about discovering and accessing the data, and then being able to generate models from those analytics that you can use to respond in your business.

We're all in a world based on data. AI has a major role to play where we can look at business processes and understand how they are operating and then drive greater automation.That's a huge focus for us -- optimizing and automating existing business processes.


We’re all in a world based on data. And we’re using it to not only look for new business opportunities but for optimizing and automating what we already have today. AI has a major role to play where we can look at business processes and understand how they are operating and then, based on analytics and AI, drive greater automation. That’s a huge focus for us as well: Not only looking at the new business opportunities but optimizing and automating existing business processes.


Listen to the podcast. Find it on iTunes. Read a full transcript or download a copy. Sponsor: IBM Storage.


You may also be interested in:

No comments:

Post a Comment