It was great to see those of you who were able to make it out to the Align IT event for Infrastructure Managers in Mississauga this week. There were quite a few folks from the Halton IT Pros group which meets just a 10 minute drive from the Microsoft office as well as my friend and former colleague, Mark Brunner, who I hadn’t seen in ages. Probably not since I left my job at Symantec for the exciting world of paint manufacturing as the IT coordinator of Valspar’s Scarborough office back in the late 90’s. But we’ve recently reconnected through the Microsoft Align IT for IT Managers LinkedIn group (yay for social media!).
I was going to put together a summary of the event for those who haven’t been able to attend (you’ve still got time – the Vancouver event on April 12 will be simulcast live online), but then I found Mark’s blog where he does a great job of doing just that. Cool…less work for me!
Here is Mark’s summary of the event which can also be found on his blog:
I just spent the day at the Mississauga Microsoft campus listening to a presentation and taking part in a discussion regarding cloud computing. Although I have read about and am familiar with the concept, this technology is new to me, and it was great to see a working environment to understand it better. I can hardly wait to roll up my sleeves and get to work planning, implementing and managing a cloud computing environment! I guess I need to find a like minded employer first. Any takers?
I was quite surprised to see these interesting stats posted on the opening slide:
- 50% of business devices are expected to be smartphones by 2014.
- 84% of organizations have a remote workforce.
- 85% of data center capacity is idle on average.
- 70% of IT budgets are spent maintaining operations.
Now, the first two don’t surprise me much, as they represent the fruits of our labors from the challenges that all IT teams have faced over the past 10 years. Mobility, connectivity everywhere, and portable device enablement. Wouldn’t it be nice if we could optimize those last 2 figures though? Imagine what your IT teams could do in the way of innovation and business enablement if we could take 20 points off each of those numbers. What wonderful new solutions would you be able to provide to your customers’ IT problems with an extra 20% of budget or processing power?
Ruth Morton presented a very brief summary of how IT as we know it has evolved. The 70′s & 80′s “hurry up and wait” mainframe days, the 90′s client/server architecture bringing power to the desktop, the 2000′s acceptance of remote and mobile access and success of the internet and web, and today’s virtual environments and sprouting implementations of cloud computing. Ruth spent some time discussing the characteristics NIST has documented as defining a cloud computing environment.
- On-demand self-service capable,
- Ubiquitous network access,
- Transparent location resource pooling,
- The ability to elastically expand and contract based on demand,
- A pay as you use, measured service model.
I was aware of the 3 types of cloud computing commonly discussed in the media, (Private, Public & Hybrid) but what really caught my attention was the definition of the 4 types of “as a Service” models within cloud computing and how responsibility and accountability could be used as the dividing factor between them. Quite logical, once the model for cloud becomes clearer. It truly is a revolutionary step in providing IT services to customers.
- Software as a Service means that the focus for the cloud is in provisioning applications. The organization is no longer responisble for maintaining the applications, platforms and infrastructure. It expects customers will consume the applications and provides efficiency as its core deliverable.
- Platform as a Service maintains responsibility for applications and data within the organization, requiring a focus on development tools. It expects IT to build solutions on the cloud platform, and provides enhanced management capabilities as a key deliverable.
- Infrastructure as a Service puts only the backbone onto the cloud, expecting the organization to maintain responsibility for the applications, data, runtime environment, middleware and operating systems. The cloud vendor would be responsible for virtualization, servers, storage and network delivery. Scalability is its key deliverable.
- The final cloud computing environment is On-Premises IT as a Service, where the entire stack is managed by the IT organization. This is common for in-house private cloud environments. Its key deliverable is control and autonomy.
Damir Bersinic, Senior Platform Advisor for Microsoft, discussed the path to delivering cloud computing services, manageing the environment with Microsoft tools, and expanded on the necessity of virualization in order to achieve the cloud model. He provided multiple demos of a live cloud environment, spoke about Microsoft, VMware, and even Linux VM provisioning, migrated and moved server VM instances seamlesly around without impacting his simulated users. Microsoft Hyper-V, System Center Virtual Machine Manager and Self Service Portal 2.0 make a pretty impressive management platform.
Three panelists, Richard Iwasa, Consulting Manager at Ideaca, Jon Platt Director of IS Shared Services at Manulife, and Joel Varty, R&D Director at Agility CMS, were on hand for the final part of the event, each discussing their individual organizations’ real world experiences with the move to the cloud. The panel members took many questions, were knowledgeable and open in their responses, and really engaged the audience members.
The implementations were each very unique, and not without issues. Joel discussed how his team managed an outage, and some of the challenges imposed by the new environment. The timeframes from proposal to production implementation completion ranged from 4 months to over a year and a half, with a significant amount of each spent on building, presenting and selling the business case.
One significant consideration that was discussed at some length was the potential of Denial of Service attacks’ changing impact from revenue depriving bandwidth consumption to a direct financial attack as the pay by the minute model is exploited to expand the bandwidth available to the “customer”. Some good advice in this regard is to plan for the potential event, build provisions into your Service Level Agreement with the cloud vendor, and set a financial cap if you can, and have a plan of action for filtering unwanted traffic. DoS remains a viable weapon in an attacker’s arsenal, and it should remain a cost of doing business consideration until we fix the underlying protocols that we rely on in Internet communications.
I really enjoyed the opportunity to be exposed to this presentation, the panel, and to meet up with a good and dear friend. Ruth and Microsoft did a fantastic job of cramming a lot of great information, clarifying complex subject matter, and looking beyond their own products when delivering the goods. Thank you to all of you for making this event possible, great job. I certainly look forward to other networking opportunities.
We’ve had great discussions at both the Montreal and Mississauga stops on the tour and I look forward to heading out to Vancouver next week and continuing the conversation. If you haven’t been able to come out, join us next Tuesday either in-person or online!