Enterprise Cloud

At a global scale, there are two major vendors that are well positioned in Enterprise I.T. within the virtualisation and OpenStack IaaS market. Of course HP and IBM are big players, but I believe their strategy somewhat splintered and certainly not “full stack”, as both VMware and Red Hat’s are. Comparing these two vendors only, there are vast differences. in particular their opposing OpenStack technology approach: VMware’s proprietary “open” integration and Red Hat’s pure Open Hybrid Cloud. These differences are even highlighted by VMware CEO Pat Gelsinger, who recently stated “On-premise cloud is a $2 trillion market … 92 percent of cloud is on-premise”. Gelsinger then went on to state that according to Gartner, this market would drop by 15% over the next 5 years. It is obvious that VMware have a lion’s share of this “on-premise cloud” market, so they’re clearly valiantly defending about $300 Billion potential loss of market. The new market entrants on the other side representing the “open” approach, are quickly building momentum with an attractive proposition.

BalticServers data center

SDN & Automation Focus

VMware is focusing on a two-pronged technology approach with NSX (Nicira) and vSphere integration into OpenStack. The amount of development contribution by VMware towards Icehouse release, particularly in Neutron is evidence enough. However what VMware and their partner network lacks is width and depth of OpenStack integration, particularly in consulting and support. VMware remain a proprietary and costly cloud solution. Although they are a strong contributor to OpenStack, they are not offering an open cloud solution. Of course VMware are about to release their vSAN solution, which takes an even more proprietary approach.

Tsunami by hokusai 19th century

Open & Full-Stack

In contrast, as both a strong Open Source and OpenStack contributor, Red Hat offers a full stack integrated proposition with key products: Red Hat Enterprise Linux (RHEL) OpenStack, CloudForms, Enterprise Virtualization, Storage and associated Services (training, consulting and support). Vital to their Open Hybrid Cloud strategy is a strong association with OpenStack and related OSS projects such as KVM, QEMU, libvirt, oVirt (potential vSphere alternative) and Gluster (including GlusterFS, bareos, Distributed Volume and HadoopVOL). In my opinion Red Hat are in an even stronger position than VMware, considering their Open Source stewardship and relationship with key OSS projects. This history and strength of cultural relationship with OpenStack project team developers will go a long way.

The key to an enterprise organisation’s successful private and hybrid cloud strategy is in utilising an Open Source solution. In leveraging Open Source in the enterprise organisation will sustain choice, mitigate vendor lock-in and minimise risk of high licensing and maintenance costs. This “open” choice will also critically enable the integration of developers and operations within an enterprise to leverage automation in a truly collaborative way. It is this vital “Lean” and innovative concept that many enterprise organisations seek and will potentially sacrifice if they maintain status quo.

 

Enterprise Strategy

The majority of Enterprise organisation’s have to-date invested in virtualisation (typically VMware), which for the enterprise has reached end of useful life with respect to realising key business benefits. We were typically having these virtualisation business benefits conversations over 9 years ago. Virtualisation alone will no longer offer business any competitive advantage. All your competitors already have implemented at least the same or deeper into their environment than you. Cloud IaaS is well into the maturity level that it is already delivering strong competitive advantage, particularly where fully integrated into product development and core business innovation. A recent study shows that “56 percent of enterprises consider cloud service adoption to be a strategic business differentiator”, which shows that 44% are blind to the reality that it IS a business differentiator. Over the next 3 years, “enterprise spending on the cloud will reach $235.1 billion”. That is solid investment in business innovation by all your competitors.

 

Napkin Sketch

Back of Napkin Assessment

There have been some shady TCO numbers quoted  by a market analyst bouncing around the Twitternet recently,  however I’m not about to complete a head-to-head TCO analysis. I will offer a “Back of Napkin Assessment”, that should highlight some salient points around costs and capabilities when it comes to cloud.

Key questions to ask:

  • What overall cost (from start of virtualisation to now)?
  • Benefits realised, opportunity/cost benefit?
  • Virtualisation business benefits now typically completed, therefore little more to gain on pure virtualisation.
  • What cost over next 5 years to maintain status quo?
  • How about cost to expand to full cloud (private/hybrid) leveraging existing vendor?
  • Not just orchestration or data centre management. Must examine the total ongoing cost.

Alternative A: maintain status quo, continue sinking more ongoing and additional cost into existing vendor = increasing licence and maintenance costs & only solves automation = locked in, expensive = choice only to be in-house virtualised, public limited, expensive and not dynamic/scalable

Alternative B: move as much as possible to public = locked in, loss control, expensive consultants, high risk

Alternative C: open hybrid cloud = leverage sunk costs (existing environment where sensible), leverage enterprise grade storage, compute, hypervisor with SLAs, no lock in, lower costs, ability to choose to be in-house virtualised, private cloud/co-lo, hybrid, public = good choice

Approximate Cost: (per socket pair) – Red Hat Enterprise Linux/OpenStack/Virtualization/CloudForms = $5K vs VMware vSphere, vCenter (no OpenStack SLA) = $10K

Recommended Enterprise approach: greenfield in R&D or other development area, particularly good at recognising benefits and cost savings. Once proven, enable strategic implementation to other significant areas of benefit.

 

Choice & Future Cost

The primary factor is choice and ongoing cost. With an open hybrid cloud, such as OpenStack you have a valuable commodity called choice. Choice of hypervisor, storage back-end, orchestration, or DevOps tools. Choice of vendor too, even the big players such as NetApp and Cisco are all over it. Service Providers are already investing. Choice is what IT decision makers want and an open cloud platform is central to this.

A pure VMware cloud solution is arguably higher in ongoing cost and limiting with future choice. I’m quite excited and look forward to even further integration into OpenStack of CloudForms, oVirt and Gluster projects, as well as further development over and above the already strong contribution by Red Hat in Heat, Marconi and Sahara OpenStack projects. All of which will open up vast opportunity for dynamic elastically scalable, orchestrated, clustered databases for Big Data workloads.

The key here is choice, once operating a stable enterprise fully supported open cloud, you then have the choice to use any other vendor or open source options in your cloud more easily. Of course there are also an ever increasing number of providers both global and local offering public and managed private/hybrid OpenStack based cloud as a product. Red Hat are further integrating with other Public Cloud providers. This all only increases your choice.

Advertisements

Focus on enabling successful business technology outcomes. I believe that Open wins over Closed and that a diverse eco-system is essential for interoperability to avoid vendor lock-in. This approach is essential for any enterprise considering hybrid cloud. Principal Solutions Architect ANZ at Nuage Networks

Tagged with: , , , , , , , , , , , , , , , ,
Posted in Uncategorized
One comment on “Enterprise Cloud
  1. Great write up! I like you point about sunk costs which leads to a requirement that can be ignored in the REHL + OpenStack solution. That’s legacy application support. In most environments that have embraced the larger VMware stack it’s to support legacy applications. Just like your OpenStack + REHL streghth observation VMware has on their said the support of the legacy data center. It will be difficult for REHL to displace VMware as the core data center virtualization platform which I think is a requirement for non-green field environments.

    Overall very good explanation of a complex challenge.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Twitter Feed
%d bloggers like this: