The recent acquisition of Nimsoft by Computer Associates, and leading venture firm New Enterprise Associates’s investment into ScienceLogic (both companies in the Availability and Performance Management of Datacenters) validate the point: datacenter and cloud computing markets are growing and a new way of managing these environments is urgently needed.

There are many write-ups on the event:

“The current market landscape consists of both products that were designed 15-20 years ago to address a different set of problems than those that end-users are experiencing today…The mid-market …underserved. Large vendors… not focusing on this market segment, and they were not able to offer solutions that were flexible and easy to use and manage, but still include all technology capabilities that end-users needed” – Bojan Simic, Trac Research http://www.trac-research.com/CANimsoft_ME.pdf

“With Nimsoft, CA can bring a tailored IT management solution to a new set of customers — emerging enterprises ($300M-$2B Revenue), emerging national markets, and the MSPs and cloud service providers that serve those markets.
But to serve these emerging enterprises, you have to do things differently
… software needs to be straightforward (and even easy) to install, use, and maintain. It needs to have a very broad but focused set of functional capabilities, stretching across a significant amount of the environments, components, and devices that an organization will want to manage, inside and outside its firewalls”, Jay Fry, CA Inc.
http://datacenterdialog.blogspot.com/2010/03/ca-and-nimsoft-because-smaller.html

I would also like to point out that these datacenter management companies becoming hot all on a sudden after 7 – 10 years in business (Nimsoft – 10 years, ScienceLogic – 7 years), shows that the market timing is now – there is a pent-up demand for better management solutions in the high growth datacenter market.

So what has all these got to do with AccelOps? In building the Integrated Datacenter and Cloud Service Management business for the last two years, many people has asked me these questions: “Why Now? Wasn’t the problem there before? Why you? Why is AccelOps better?”. After answering the same questions in many emails, calls to customers, market analysts, investors and employees, I thought it would be better to just put down my thoughts here…

So let’s get to the ‘Why Now?’ question.

In order to answer that, let’s look at what is normally called ‘the compelling event’ in the sales practice. A compelling event consists of “what is pain point, and is it time sensitive”?

Driven by outsourcing, consolidation and cost optimization, the datacenters in recent years is going through a lot of changes: the proliferated use of VMs, VMware’s reference storage architecture and unification of storage and networking.

Today’s datacenter is definitely more complex and there are many moving parts – it is not the same datacenter that we used to know 10, 20 years ago. Things are more fluid now: resource optimization is causing Virtual machines to move between physical machines and dynamically connect to different virtual switch ports at different times, applications are dynamically load balanced across different virtual machines in a cluster, users access the network from different entry points (mobile, vpn, wired ports).

In the old days, server environments used to be static. The presentation layer (GUI), the computation layer (servers) and the storage layer were relatively independent of each other. But that is not the case anymore: VM images are stored in the remote shared store and moved around by the hypervisor, data are computed in the servers and shipped across high speed networks to remote stores, networking and storage traffic are sharing the same switch. The storage layer is becoming a very integral part of the computation layer. Without much need to say, same is true for the network layer…

The line between applications, servers, network and storage are becoming blurred. Cisco’s attempt to unify computing, networking and storage by entering the server business (UCS), HP getting into the high-end enterprise switching market (via 3Com acquisition), Oracle buying Sun signify just that.

So that is the pain point: complex and dynamic environment and how do you manage it?

Now, let’s get to the second attribute of the ‘Compelling Event’: time sensitivity.

This complex data center environment happens to support business critical business applications and services. This is especially true these days for the service providers and cloud providers, where their top line (revenue) comes directly from the datacenter operations. There is a SLA tag to business applications. When I was working in AT&T and Bell Labs, 99.999% was the only thing for the telephone network. Now 99.999% is the requirement for the business applications and the data network and is the critical differentiator for service providers! With that level of SLA attached to the complex datacenter, problems need to be resolved fast, and problems need to be predicted and avoided.

By the way, not an after thought, what about the security threats to the datacenter and cloud? How do you keep it secured and the data protected? Do you have a situation where you are debugging a network slowness issue, while wondering whether it could be the result of an access control violation, an unauthorized change, an unpatched server in the test network generating lots of network traffic, or a server performance problem? How do you quickly find out who has access to the information, while you may not have the permission to do so? What about compliance? It is part of the service delivery too! No one is going to dispute that when a security threat brings down a datacenter network, it has nothing to do with SLA…

So how do we solve this new problem? In the next blog, I will write about the ‘how’ part, which is ‘why you?’, ‘Why AccelOps?’…

Thanks,

Imin

Tags: Cloud Management, Datacenter Management