BPOS 2010 and “Superset”

I was reading one of these pretty generic blah-like articles on BPOS (Microsoft’s exchange and SharePoint in the cloud) and thankfully waded through to the end:

In terms of other near-term deliverables, Microsoft is commiting to provide in BPOS v.Next native PowerShell scripting via a PowerShell endpoint build on PowerShell Version 2. Authentication will be done through Online IDs, with a single credential being able to be used for both PowerShell and the portal.Keane echoed the message other Microsoft execs have been voicing at TechEd this week: Cloud capabilities, over time, will  become a superset of what is available on-premises. Currently, the reverse is true, and Microsoft’s Online services offer a subset of the functions available in the software equivalents of each product.

The notion that the cloud will provide more capability than on-premise is new to me. I wonder how true that is going to be in the end.  It feels counterintuitive to me.  I totally get the idea that a lot of companies will move stuff to the cloud (or start off in the cloud) but I normally think they do it because the pro’s (easier admin, SLAs, etc) far outweigh the cons (reduced functionality). 

I’m having a little bit of a hard time believing that cloud offerings will exceed on-prem capabilities.  Multi-tenant is hard and seems like it must force compromises in order to provide good SLA and ease of use…

I’ll probably be eating my words on this.  I remember thinking that no one could possibly need more than 650 mb of data and therefore, the CD was never going to be improved upon.


Subscribe to my blog.

Follow me on Twitter at http://www.twitter.com/pagalvin

2 thoughts on “BPOS 2010 and “Superset”

  1. Jeff Becraft

    Hey, Paul. I always love to read your stuff.

    I saw that line too, and thought it was just as provocative… But, I believe they stopped short of saying multi-tenant. So, my take on it is that the superset would apply where the software is “in the cloud,” and leveraging multi-tenant infrastructure perhaps, but not multi-tenant application layer, at least for now.

    And in that sense, I tend to agree with the statement, in that the “cloud,” meaning a hosted and managed solution does now and/or will offer a superset of on-premises capabilities.

    For example, my customers get a dedicated, full-featured SharePoint that we allow them to customize (equal to on-premise), along with a 99.9% uptime guarantee SLA, and performance SLA, on a solution that leverages our shared infrastructure platform to optimize costs and enable highly dynamic scaling, so they pay for only what they use of bandwidth, storage, computing capacity, licenses, etc.

    Plus, we have tremendous networking capabilities for connecting the system to customer’s users around the world, so they don’t have to give up as much as they might think in terms of network performance. On top of that, they can leverage our ICDS content delivery network and/or optionally apply WAN optimization (Riverbed, for example) to the solution to boost network performance even further. The option for remote BLOB storage is there as well, and that can be pointed to our Storage as a Service platform to get cheap storage for older or non-critical data.

    Then you have to consider that many companies lack state of the art data centers in multiple locations around the world. A provider who can make that available easily offers a tremendous value for geo-distributed corporations.

    When you put all that together, it is a lot more than most CIOs can match on-premise.

    1. Paul Galvin Post author

      I think you make a great case for the cloud there.

      I think that companies still worry about:
      – Monthly costs
      – Customer service
      – Security of data
      – Integrating their cloud data with their on-prem systems. For instance, custom SQL databases (an issue I’m working out with a client right now).
      – Other stuff no doubt 🙂

      To a small extent, this is sort of like the old javascript / cookie issue. We actually had design requirements that said that our web pages needed to work reasonably well if the user turned off cookies and disallowed javascript and we’d actualy advertise that our stuff worked without the need for these “potential security problems”. These days, no one even bothers to support users that won’t accept cookies or disable javascript on their browser.


Leave a Reply

Your email address will not be published. Required fields are marked *